Physiol Rev 89: 777–798, 2009;
doi:10.1152/physrev.00026.2008.
LKB1 and AMPK Family Signaling: The Intimate Link Between
Cell Polarity and Energy Metabolism
MARNIX JANSEN, JEAN PAUL TEN KLOOSTER, G. JOHAN OFFERHAUS, AND HANS CLEVERS
Hubrecht Institute, Developmental Biology and Stem Cell Research, and Department of Pathology, University
Medical Centre, Utrecht; and Department of Pathology, Academic Medical Centre, Amsterdam, The Netherlands
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
I. Introduction
II. Peutz-Jeghers Syndrome
A. Historical account
B. Clinical characteristics
C. Cancer risk and surveillance
III. Cloning of the LKB1 Tumor Suppressor Gene
IV. Mouse Models
A. Embryogenesis
B. Polyp development
C. Tumor models
D. LKB1 dosage influences cancer susceptibility
V. Molecular Characteristics of LKB1
VI. LKB1-STRAD-MO25 Complex
A. STRAD isoforms
B. MO25 isoforms
C. Complex formation and LKB1 activation
VII. The Metabolic AMPK Module Is a Downstream Target of LKB1
A. Physiological regulation
B. LKB1 is an AMPK kinase
C. LKB1 exhibits isoform specificity towards AMPK
D. mTOR module
E. Clinical applications
VIII. LKB1 and Cellular Polarization
A. A family of Par proteins regulates cellular polarity
B. LKB1 in epithelial polarity
C. LKB1 in neuronal polarity
D. Role of LKB1 in asymmetric cell division
IX. Conclusion
778
778
778
778
780
781
781
781
781
782
782
783
783
783
784
784
784
784
785
785
786
787
787
787
788
790
790
793
Jansen M, ten Klooster JP, Offerhaus GJ, Clevers H. LKB1 and AMPK Family Signaling: The Intimate Link
Between Cell Polarity and Energy Metabolism. Physiol Rev 89: 777–798, 2009; doi:10.1152/physrev.00026.2008.—
Research on the LKB1 tumor suppressor protein mutated in cancer-prone Peutz-Jeghers patients has continued
at a feverish pace following exciting developments linking energy metabolism and cancer development. This
review summarizes the current state of research on the LKB1 tumor suppressor. The weight of the evidence
currently indicates an evolutionary conserved role for the protein in the regulation of various aspects of cellular
polarity and energy metabolism. We focus on studies examining the concept that both cellular polarity and
energy metabolism are regulated through the conserved LKB1-AMPK signal transduction pathway. Recent
studies from a variety of model organisms have given new insight into the mechanism of polyp development and
cancer formation in Peutz-Jeghers patients and the role of LKB1 mutation in sporadic tumorigenesis. Conditional LKB1 mouse models have outlined a tissue-dependent context for pathway activation and suggest that
LKB1 may affect different AMPK isoforms independently. Elucidation of the molecular mechanism responsible
for Peutz-Jeghers syndrome will undoubtedly reveal important insight into cancer development in the larger
population.
www.prv.org
0031-9333/09 $18.00 Copyright © 2009 the American Physiological Society
777
778
JANSEN ET AL.
I. INTRODUCTION
II. PEUTZ-JEGHERS SYNDROME
A. Historical Account
Peutz-Jeghers is a rare cancer predisposition syndrome characterized by the development of gastrointestinal polyps and mucocutaneous pigmentation abnormalities. The syndrome was first described in 1921 by the
Dutch physician Dr Johannes Peutz working in The Hague
(23, 80, 93). He documented the case of a 15-yr-old boy
that had been surgically treated for a small bowel intus-
B. Clinical Characteristics
The first sign of PJS is commonly a distinctive pigmentation observed around the lips, oral mucosa, genitalia, or
palmar surfaces appearing early in childhood (Fig. 1) (60,
80). However, the polyps constitute the main clinical
symptom. These may develop throughout the gastrointestinal tract, grow to large sizes, and cause occlusions, pain,
and gastrointestinal bleeding with anemia. Polyps in PJS
appear to predominate in the small intestine. A recent
study involving repeated gastroscopies in PJS patients
showed the presence of hundreds of smaller polyps in the
stomach of all analyzed patients (123). These small, nascent PJS polyps lacked the prominent smooth muscle
stalk typical of larger PJS polyps, but did show signs of a
hyperproliferative epithelium, and were classified as hyperplastic polyps, accordingly. These observations suggest that counting polyps limited to those that come to
clinical attention might misrepresent actual numbers and
FIG. 1. Clinical characteristics of Peutz-Jeghers syndrome (PJS). A: perioral pigmentation observed in PJS patient. B: high-power
photomicrograph of hematoxylin and eosin staining of polyp removed from PJS patient showing characteristic smooth muscle proliferation;
dysplasia is absent. Inset shows characteristic branching of smooth muscle. C: gross macroscopy of PJS polyp shown in B. Note characteristic
cauliflower-like appearance of the mature PJS polyp. A video recording of endoscopic PJS polyp removal in a PJS patient may be found at
http://www.youtube.com/watch?v⫽BM8j_Cx_kNA.
Physiol Rev • VOL
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
In recent years, research on the LKB1 tumor suppressor protein has intensified at a staggering pace. Historically, interest in the protein was first aroused in 1998 after
it was found that germline inactivating mutations cause
the rare cancer-prone Peutz-Jeghers syndrome (PJS). Although initial studies appeared to suggest a limited role
for the protein in sporadic malignant tumor progression,
it has now become evident that LKB1 is frequently targeted for inactivation during sporadic lung cancer transformation. In addition, studies now suggest a role for
LKB1 in metabolic regulation through the well-known
AMP-activated protein kinase (AMPK) module. For this
reason, LKB1 has now been implicated in a vast range of
clinical conditions ranging from diabetes mellitus to pulmonary cancer. As these conditions represent primary
causes of morbidity in the Western world, LKB1 has become a formidable target for drug therapy. In keeping
with this range of physiological processes and associated
clinical disorders, LKB1 is now linked to a broad plethora
of downstream targets. This review aims to provide an
overview of the current state of research on the LKB1
tumor suppressor protein.
susception. On rectoscopy, Dr. Peutz noted multiple rectal polyps in this patient. Further investigation in siblings
led to the identification of similar rectal polyps. All siblings affected by gastrointestinal polyps similarly suffered
from nasal polyps and perioral hyperpigmentation. The
syndrome was further characterized by Dr. Harold Jeghers in 1949, who recognized that “a single pleiotropic gene
was responsible for both characteristics, the polyps and
the spots.” PJS is an autosomal dominant cancer-prone
gastrointestinal polyposis disorder. The disorder presents
in 1 per 50,000 to 1 per 200,000 newborns and is caused by
germline mutation in the LKB1 gene. Due to the rarity of
the syndrome, clinical accounts of pedigrees afflicted by
the syndrome are rare, and accurate documentation of the
increased cancer-risk is therefore complex. Based on large
cohort studies, patients are estimated to display an 18-fold
increased cancer risk over the normal population (43).
LKB1-AMPK FAMILY SIGNALING
the key question of how malignant derailment in PJS
patients models malignant tumor progression in sporadic
patients, i.e., cancer patients not affected by PJS. Analogous to the lessons learned from the stepwise tumor
progression that occurs in familial adenomatosis polyposis (FAP) patients, it is hoped that neoplastic transformation in PJS patients similarly models events during malignant tumor progression, which can then be extrapolated
to the public at large. A firm understanding of the histopathological mode of cancer initiation in PJS patients is
critical to begin to unravel the role of LKB1, and the
consequences of its loss, during sporadic tumor progression. Because of their marked presentation, the gastrointestinal polyps in PJS constitute a prime suspect, but do
they really harbor an underlying risk for malignant transformation in spite of their benign appearance? Taking the
above-mentioned case of a conventional adenomatous
polyp displaying a prominent PJS-like core of arborizing
smooth muscle as an example, it would on histopathological grounds alone not be possible to discern whether one
is observing an adenomatous polyp that secondarily acquired a PJS-like smooth muscle stalk or, alternatively, a
PJS polyp that has undergone adenomatous transformation (Fig. 2). Historically, case reports have described
dysplastic adenomatous transformation occurring in a
supposedly preexistent “hamartomatous” PJS polyp (83).
However, dysplasia occurring in classical PJS polyps appears to be a very uncommon phenomenon. In series from
the renowned familial polyposis registry at St Mark’s hospital, a review of 491 polyps showed no evidence of
dysplasia in any of them (106). Moreover, follow-up over
FIG. 2. PJS-like smooth muscle proliferation in a conventional non-PJS adenomatous polyp. A: gross macroscopy of a pedunculated colonic
polyp (indicated by * in A) in a 67-yr-old patient not affected by PJS. A right hemicolectomy was performed following the diagnosis of colon cancer
in the ascending colon in this patient; the polyp shown here was found incidentally in the cecum. B: low-power photomicrograph of the polyp shown
in A. The polyp has a tubulovillous architecture characterized by long fingerlike glands emanating from the stalk of the polyp. Note the slender
extensions of the smooth muscle core into the tips of the villous projections (indicated by * in C). This arborizing core of smooth muscle resembles
the histology shown in Fig. 1B for the polyp removed from a PJS patient. C: unlike PJS polyps, which are characterized by a nondysplastic
hyperproliferative epithelial lining, this polyp has a dysplastic epithelial covering classified as low-grade dysplasia. The boxed area in B is shown
at higher power in C. Polyps removed from PJS patients have been reported to contain dysplastic foci, which have been taken as evidence of
malignant potential of PJS polyps in spite of their benign histological appearance (see sect. IVB on models of polyp development in PJS). However,
conventional preneoplastic adenomatous polyps arising in PJS patients may similarly have features such as a prominent smooth muscle core along
the stepwise progression towards an invasive carcinoma, as shown in this example. PJS polyps as shown in Fig. 1B may therefore be an
epiphenomenon to the malignant predisposition in PJS and not preneoplastic per se.
Physiol Rev • VOL
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
distribution of polyps. Polyps in PJS have also been described at locations outside of the gastrointestinal tract
such as the nasopharynx, gallbladder, trachea, and urogenital tract.
The large gastrointestinal polyps typically encountered in PJS, which may develop from the smaller hyperplastic lesions, are pedunculated polyps with nondysplastic overlying epithelium. A characteristic feature of mature PJS polyps is the prominent core of arborizing
smooth muscle, which extends into the head of the polyp.
Smooth muscle proliferation appears to accompany epithelial hyperproliferation during early stages of polyp development (54). It is important to realize that smooth
muscle proliferation is not a histopathological feature
specific for the PJS polyp. For example, PJS-like smooth
muscle proliferation can also be observed in conventional
sporadic adenomatous polyps. In this instance, the smooth
muscle proliferation develops secondary to mechanical
insults (due to intestinal peristalsis) during adenomatous
polyp growth (54). Furthermore, other conditions unrelated to PJS such as mucosal prolapse syndrome may
present with histological features similar to those observed in PJS polyps as well (54, 107). Thus it is unclear
whether the observed smooth muscle proliferation is
causally involved in polyp initiation or, alternatively,
whether it develops as an epiphenomenon to accommodate for epithelial hyperproliferation.
The potential for malignant derailment of PJS polyps
and, specifically, whether cancer in PJS patients arises
from the nondysplastic epithelium covering PJS polyps
remains a hotly debated subject. This debate centers on
779
780
JANSEN ET AL.
Physiol Rev • VOL
classification of PJS polyps thus awaits elucidation of the
sequence of events during PJS polyp initiation and the
pathophysiological mechanism responsible for PJS polyp
development.
C. Cancer Risk and Surveillance
Mechanical problems due to polyp development in
PJS dominate the first two decades of life, but with advancing age intestinal and extraintestinal cancer development becomes a major clinical concern. The cancer spectrum consists mostly of gastrointestinal cancers (colorectal and pancreatic) (43, 74). However, an increased risk
for extraintestinal cancers has also been noted, and the
risk of breast cancer in female PJS patients is comparable
to that associated with either BRCA1 or BRCA2 mutations (43). Thus, compared with other cancer-prone conditions, PJS patients display an unusually wide tumor
spectrum. In addition, PJS patients are at increased risk
for distinctive benign tumors of the genital tract such as
sex cord tumors with annular tubules (SCTATs) in affected female patients. SCTATs in PJS are often bilateral,
multifocal, and benign and have been described as characteristic for the condition (135). SCTATs in patients not
affected by PJS are typically unilateral and often display a
malignant clinical course, in contrast to its benign character in PJS. Even though the absolute risk for the development of these gonadal tumors is only modestly increased, their occurrence is noteworthy as it may provide
an important clue to the underlying mechanism of PJS
polyp development and neoplastic transformation in PJS
(see below).
The range of organs affected in PJS translates into a
cumbersome follow-up strategy, and current recommendations include biannual upper endoscopy with polypectomy, colonoscopy with polypectomy, and small bowel
X-ray series (Table 1) (36). However, a rational basis for
surveillance strategies in these patients is sorely lacking.
A recent analysis of the psychosocial impact of PJS in
affected patients has shown that a diagnosis of PJS affects
TABLE 1. Cancer screening recommendations
for Peutz-Jeghers syndrome patients
Screened Organ
Starting
Age, yr
Interval, yr
Tests
25
10
2
2
30
20
20
20
10
1–2
2
1
1
1r
Colonoscopy
Endoscopy, small bowel
barium X-ray series
Endoscopic ultrasound
Mammography
Transvaginal ultrasound
Pap smear
Physical exam, ultrasound
Colon
Proximal GI,
small intestine
Pancreas
Breast
Uterus
Cervix
Testicles
GI, gastrointestinal tract. Data from Giardiello and Trimbath (36).
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
45 years of 48 patients at the Mayo Clinic failed to reveal
dysplastic change in any polyp as well (75). More recently,
we have proposed that these historical accounts of adenomatous transformation occurring in PJS polyps might
be explained by a blurring of the order of events; that is,
these lesions might likely have arisen due to a conventional adenoma arising in a PJS patient that secondarily
developed a PJS-like smooth muscle stalk, akin to the
situation encountered in sporadic patients, as described
above (54). This is addressed in greater detail in section
IVB on models of polyp development.
In view of the fact that the malignant potential of PJS
polyps, and by inference the focus of cancer initiation in
PJS patients, remains unclear, proper classification of PJS
polyps is a relevant issue. Traditionally, mature PJS polyps have been classified as hamartomas, which signifies a
nondysplastic overgrowth of all tissue layers native to the
site of origin of the lesion in equal proportion. It is important to recall, however, that the “hamartoma” designation for PJS polyps is not based on a pathophysiological
rationale at present. The hamartoma classification may
therefore be misleading (and may ultimately prove to be a
misnomer), since the actual etiological mechanism of
polyp development remains unclear. This consideration is
also relevant with regard to a number of studies that have
appeared in literature mechanistically linking the socalled hamartoma syndromes PJS, tuberous sclerosis, and
Cowden’s disease (24). Irrespective of whether a definitive molecular link can be established in future (see sect.
VIID on the mTOR module), from a clinical point of view it
appears artificial to lump these syndromes together since
they share little more clinical overlap than the hamartoma
designation per se. Further research examining the histological architecture of PJS polyps in closer detail is
therefore warranted. For example, even though the hamartoma designation mandates that the epithelial covering of
PJS polyps would display all lines of differentiation normally observed at the site of origin of a PJS polyp in a
normal ratio, earlier studies have pointed out that with
regard to the epithelial covering of PJS polyps, some
differentiated epithelial lineages may be lost at the expense of other lineages or in favor of a more immature
phenotype altogether (46). The net shift that occurs towards an immature phenotype is not to be interpreted as
a sign of premalignant potential. For these reasons, and to
avoid clouding the debate, it appears prudent to refer to
the gastrointestinal lesions in PJS as “polyps” or “PJS
polyps” rather than hamartomas, since the latter would
imply an etiological understanding that is currently not
justified. Most importantly, because the malignant potential of PJS polyps, i.e., those polyps demonstrating a
prominent smooth muscle core with normal nondysplastic overlying epithelium, remains uncertain, PJS polyps
should be clearly separated from other (pre)neoplastic
lesions occurring in Lkb1 mouse models (50). Accurate
LKB1-AMPK FAMILY SIGNALING
many important life decisions in these patients even
though physically patients do not feel impacted compared
with the general population (131).
III. CLONING OF THE LKB1 TUMOR
SUPPRESSOR GENE
IV. MOUSE MODELS
A. Embryogenesis
In an attempt to define the physiological function of
LKB1 in the mammalian setting, several laboratories have
Physiol Rev • VOL
created mice carrying mutations targeted to the Lkb1
locus. Data show that functional Lkb1 is required for
normal mouse embryogenesis. Lkb1 null mice do not
survive beyond embryonic day (E) 10.5 and show several
developmental abnormalities including impaired vascular
development and shortened body axes (134). A role for
increased vascular epidermal growth factor (VEGF) production in relation to the vascular abnormalities in these
embryos has been discussed (134). Lkb1 ⫺/⫺ mice survive to a stage of embryogenesis resembling approximately E7.5, passing through major developmental events
such as implantation and gastrulation. The fact that Lkb1
null mice survive to a late stage of embryogenesis stands
in clear contrast to the consequences of targeted biallelic
Lkb1 recombination at later stages of development. Here,
loss of Lkb1 in, for example, neural or pancreatic progenitors has clear cell-autonomous deleterious effects associated with a loss of cell polarity (12, 48). This puzzling
context-dependent effect of the consequences of Lkb1
removal remains hitherto unexplained.
B. Polyp Development
Lkb1 heterozygous mice have proven instrumental in
modeling the role of Lkb1 during PJS polyp development
and tumor progression in the mammalian setting. Mouse
models established in different laboratories have shown
that biallelic Lkb1 inactivation is not required for polyp
development (9, 57, 85, 92, 127). Lkb1 ⫹/⫺ mice develop
polyps that histopathologically mirror their human counterpart. Polyps retain the wild-type Lkb1 allele, showing
that Lkb1 is haploinsufficient for suppression of polyposis. Polyps are first detected at around 5 mo and cause
premature lethality from 8 mo onwards, presumably due
to intestinal blockage. Polyps are particularly prominent
in the stomach, in line with recent reports on polyposis
in the human stomach (59, 123). Recent data show that
neither a p53 ⫹/⫺ nor a p53 ⫺/⫺ background allows
for malignant transformation of murine Lkb1 ⫹/⫺ polyps (116, 127). Moreover, while dimethylbenz[a]anthracene(DMBA)-treated Lkb1 ⫹/⫺ mice (in a systemic
carcinogen protocol) develop squamous cell carcinomas of the lung and skin, the number and histopathological features of the polyps remain unchanged (39). These
observations suggest that murine PJS polyps carry little
malignant potential.
The issue of polyp development in PJS patients has
been surrounded by a long-standing debate concerning
the potential for malignant degeneration of PJS polyps.
Regardless of whether the polyps are ultimately deemed
to lack malignant potential, it is clear that PJS polyp
development and neoplastic transformation must both be
accounted for by the same genetic mechanism, underscoring the need to elucidate polyp development in PJS.
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
The LKB1 gene was cloned after comparative genomic
hybridization (CGH) analysis on PJS polyps, which reportedly showed loss at chromosome 19p13 in a number
of polyps removed from patients (46). Subsequent studies
involving linkage analysis in PJS pedigrees confirmed
linkage to this region. This inspired a commendable
search for mutations in genes located in this genomic
region, which allowed for the identification of mutations
in the LKB1 gene (45, 55). Initial studies suggested locus
heterogeneity, since LKB1 mutations were not detected
in a significant number of cases. However, with the advent
of techniques to identify large genomic deletions, it became clear that germline LKB1 inactivation is the central
culprit (5, 29, 44). Indeed, sequencing analysis of interacting partners of LKB1 or potential downstream effectors
has not shown any mutations (2, 29, 31). In a remarkable
twist, a recent study documented an activating germline
mutation in the smooth muscle myosin gene (MYH11) in
a patient displaying features consistent with PJS (3). The
gene was analyzed following the zebrafish meltdown mutant, which carries an activating germline mutation in the
zebrafish MYH11 homolog. This mutant is reported to
demonstrate intestinal abnormalities resembling those of
PJS and juvenile polyposis syndrome. Future studies, for
example, involving murine knock-in models designed to
express an activating MYH11 mutation, may clarify whether
this mutation indeed predisposes to PJS polyposis.
The identification of the LKB1 tumor suppressor gene
afflicted in PJS patients prompted the analysis of LKB1
sequence changes in sporadic patients. Initial studies
showed, perhaps surprisingly, that LKB1 was only infrequently biallelically targeted in sporadic tumorigenesis (e.g.,
⬍5% of sporadic colorectal carcinomas show biallelic inactivation) (101, 122). In contrast, recent research now shows
that LKB1 is frequently targeted for inactivation in sporadic
pulmonary cancers, in particular its most frequent subtype
adenocarcinoma (56, 79, 97). Since lung cancer is the most
frequent cause of cancer mortality in the Western world, this
clearly validates research on rare cancer predisposition syndromes such as PJS.
781
782
JANSEN ET AL.
Physiol Rev • VOL
C. Tumor Models
Models investigating the role of murine Lkb1 during
malignant tumor progression have firmly established its
tumor suppressive qualities. One study has reported high
numbers of hepatocellular carcinoma (⬎70%) in aged
Lkb1 ⫹/⫺ mice, but this has not been confirmed in other
models, nor is hepatocellular carcinoma part of the PJS
tumor spectrum (87). More recently, female Lkb1 ⫹/⫺
mice have been reported to spontaneously develop welldifferentiated endometrial carcinomas with long latency,
which are potentially analogous to adenoma malignum in
female PJS patients (27). Conditional biallelic Lkb1 inactivation has been shown to result in neoplastic transformation in a number of tissues including lung (56), epidermis (39), pancreas (48), endometrial lining (27), and prostate (90). It is possible that the relative paucity of
neoplastic transformation in Lkb1 ⫹/⫺ mice not bearing conditional alleles is due to the fact that these mice
often develop fatal PJS polyps at an early age. Crosses
of Lkb1 ⫹/⫺ and p53 ⫹/⫺ and p53 ⫺/⫺ mice have
shown that these mice are highly tumor-prone, show an
accelerated tumor-onset, and typically develop neoplasms part of the p53 tumor spectrum (sarcomas,
lymphomas) (116, 127).
D. LKB1 Dosage Influences Cancer Susceptibility
By definition, LKB1 is haploinsufficient for the development of PJS, given that hemizygosity leads to characteristic phenotypes including gastrointestinal and nasal
polyposis (61). With respect to its role as a tumor suppressor, recent studies now indicate that LKB1 dosage
may critically influence cancer susceptibility. A most recent study investigating the role of Lkb1 in pulmonary
carcinogenesis has shown that monoallelic Lkb1 inactivation cooperates with K-Ras mutation. Here, Lkb1 hemizygosity facilitated transformation and tumor metastasis,
whereas biallelic inactivation of a conditional Lkb1 allele
was associated with shorter latency and a different histological spectrum (56). Critically, Lkb1 inactivation altered the resulting spectrum of tumor histologies; mice
carrying conditional Lkb1 alleles develop adenocarcinoma, squamous carcinoma, and large cell carcinoma,
whereas tumors in other genetic backgrounds are strictly
adenocarcinomas. In mice carrying a single conditional
Lkb1 allele, the wild-type allele was apparently retained.
This is similar to what has been observed on several
occasions in murine models in other studies (27, 127) as
well as in material derived from PJS patients (61, 101).
Loss of the wild-type allele may thus afford additional
selective advantages, but Lkb1 hemizygosity is clearly not
neutral. Other well-known tumor suppressors such as
PTEN have also demonstrated a context-dependent out-
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
The demonstration of loss of the wild-type LKB1 allele in
human PJS polyps by CGH analysis initially inspired the
proposal of the hamartoma-carcinoma sequence as the
mechanism of tumor formation in PJS (17). This concept
implied that PJS polyps, or hamartomas, are prone to
malignant degeneration, in spite of their benign appearance, and that LKB1 acted as a classical tumor suppressor requiring biallelic inactivation for phenotypic expression. Recently, we proposed that PJS polyps are not premalignant and are a signpost to the malignant condition,
not its direct obligate histological precursor (54). Polyps
both in patients (30, 47) as well as in murine models (9,
57, 85, 92, 127) retain the wild-type allele. Moreover,
polyps removed from patients are polyclonal (30). These
observations do not accord with a clonal expansion scenario and argue in favor of a lack of malignant potential of
PJS polyps.
Interestingly, a recent study involving Lkb1 recombination limited to the mesenchymal compartment suggests
that polyp development results from defective epithelialmesenchymal cross-talk, implying a “landscaper” scenario
for PJS tumorigenesis (64). Irrespective of whether one or
both “floxed” Lkb1 alleles were deleted upon expression
of a Cre transgene driven by the smooth muscle actin
(Sm22) promoter, polyposis ensued that was histopathologically indistuingishable from the polyps encountered in
previous nonconditional Lkb1 ⫹/⫺ models (59). Due to
reasons that are currently unclear, polyp burden was
lower in these mice carrying conditional Lkb1 alleles
compared with polyp burden in previous nonconditional
models. The mechanism underlying polyp formation in
this model has been suggested to involve a lack of secretion of growth and migrational stimuli such as transforming growth factor (TGF) and platelet-derived growth factor (PDGF) by Lkb1 ⫹/⫺ cells in the mesenchyma (59,
124). In a second study employing a similar strategy,
endothelium-specific Lkb1 removal using the Tie1-Cre
allele revealed that Lkb1 ⫺/⫺ yolk sac vessels in murine
embryos appeared dilated and distorted. This defect in
vascular remodeling was traced to a lack of recruitment
of vascular smooth muscle cells due to defective TGF
signaling by the Lkb1-negative endothelium. A landscaper role for LKB1 in PJS suggests that PJS polyp
development and malignant transformation occur due
to inappropriate proliferative stimulation of the overlying epithelial compartment by the surrounding mesenchyma. Studies investigating a landscaper role for the
LKB1 tumor suppressor should be predicated on the
demonstration of the presence of both wild-type alleles
in the overlying epithelium. The implications of this
novel concept with regard to the debate surrounding
the malignant potential of PJS polyps and the role of
LKB1 in sporadic tumor development remain currently
unclear.
LKB1-AMPK FAMILY SIGNALING
come with regard to gene dosage (96). This implies that
LKB1 may not adhere to the classical premise of tumor
suppression, wherein both copies of a given tumor suppressor gene must be affected in order for malignant
transformation to arise. Future research will be aimed at
resolving this issue (102).
V. MOLECULAR CHARACTERISTICS OF LKB1
Physiol Rev • VOL
rescue activity (78). These data indicate that the extreme
COOH terminus of LKB1 is important for its function.
Prenylation of endogenous LKB1 and the functional consequence of this modification have thus far not been
investigated in a mammalian setting.
Recently, a novel shorter splice variant of LKB1 has
been described. This shorter variant carries a unique 39residue COOH-terminal sequence lacking some of the
known phosphorylation and farnesylation sites described
above, which may imply alternative posttranslational regulation of this variant. Although the exact expression
pattern needs to be further characterized, depletion of
this shorter splice variant in mice revealed a role in
spermiogenesis (32, 120).
VI. LKB1-STRAD-MO25 COMPLEX
Although LKB1 had previously been linked to a plethora of cellular processes, little was known about its physiological regulation. Major insight into its regulation came
with the observation that the Ste20 adaptor protein
STRAD binds LKB1, which stabilizes and activates LKB1
(6). Further biochemical studies on the complex led to the
identification of the scaffolding protein MO25 as a third
component of the trimeric LKB1-STRAD-MO25 complex
(18, 22).
A. STRAD Isoforms
The human genome encodes two isoforms of STRAD
termed STRAD␣ and STRAD␤ (1). Both isoforms have
been classified as “pseudokinases” as they lack several
key catalytic residues. Out of 518 protein kinases encoded by the human genome, 48 have been classified as
pseudokinases (77). Sequence analyses of these pseudokinases indicate that they lack at least one of three motifs in
the catalytic domain that are essential for catalysis.
STRAD carries mutations in the third Asp-Phe-Gly (DFG)
motif in subdomain VII of the kinase domain. The aspartic
acid in this motif binds the Mg2⫹ that coordinate the ␤and ␥-phosphates of ATP in the ATP-binding cleft of
nonmutated, active kinase domains (20). Structural analysis has also shown that STRAD carries many inactivating
mutations at its catalytic domain (82), and mutations
aimed at restoring the catalytic activity of STRAD␣ by
mutating residues back to those found in active kinases
failed to reactivate STRAD␣ (20). Indeed, STRAD␣ does
not autophosphorylate or phosphorylate any substrate
tested (6). Curiously, although STRAD is an inactive
pseudokinase, it still possesses several conserved motifs
found in active protein kinases including the Gly-rich
P-loop motif required for ATP binding. Indeed, STRAD
does interact with ATP and ADP with high affinity (Kd
30 –100 ␮M) (20). The functional significance of this is
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
The LKB1 gene is ubiquitously expressed in both
adult and fetal tissues (1, 60). The gene spans 23 kb and is
composed of 10 exons, 9 of which are coding. It is transcribed in the telomere-to-centromere direction and encodes a highly conserved 50-kDa serine/threonine kinase.
Alignment studies show that LKB1 has no close relative
within the human genome, which similarly pertains to the
mouse, fly, and worm genome. The catalytic domain
shows the highest degree of conservation. Interestingly,
inactivating mutations both from PJS patients and sporadic cases are not restricted to the kinase domain (1, 60).
Human LKB1 is a 436-amino acid protein that consists of
an NH2-terminal noncatalytic domain, a kinase domain
(residues 49-309) that is most similar to the SNF1/AMPactivated protein kinase family, and a putative COOHterminal regulatory domain. Endogenous and transfected
LKB1 is present both in the nucleus and cytoplasm of
cells. Two potential nuclear localization signals (NLS) are
located at amino acids 38 – 43 and amino acids 81– 84 (1,
108). A number of residues on LKB1 are either autophosphorylated (Thr-185, Thr-189, Thr-336, and Ser-404) or
phosphorylated by upstream kinases (Ser-31, Ser-325,
Thr-366, and Ser-431), and the residues surrounding these
sites are highly conserved in Drosophila, Xenopus, and
mammalian LKB1. Thus far, there are no reports showing
that mutating these phosphorylation sites to Ala or Glu
has an effect on kinase activity (18, 99, 100). However,
changing Ser-431 into an Ala resulted in an LKB1 mutant,
which is retained in the nucleus (109), suggesting that
protein kinase C (PKC)-␨, protein kinase A (PKA), or
ribosomal protein S6 kinase (RSK) could regulate the
localization of LKB1 by phosphorylating Ser-431 (99, 100,
109) and thereby stimulate the active transport of LKB1
out of the nucleus (33).
The COOH terminus of LKB1 contains a so-called
CAAX-box, a consensus sequence for prenylation by addition of a farnesyl group allowing for plasma membrane
insertion. LKB1 is indeed prenylated in cultured cells at
Cys-433 (26, 100) and has, at least in invertebrate systems,
been shown to be targeted to the plasma membrane (78,
126). It is interesting to note that naturally occurring PJS
mutations include stop mutations that would prevent
translation of the last 20 amino acids (1). Additionally,
point mutation of the residue homologous to Cys-433 in
Drosophila constitutes an allele with severely reduced
783
784
JANSEN ET AL.
MO25␣ reduces the amount of endogenous STRAD␣ associated with LKB1 (18). Note, however, that most of
these cell culture experiments involve overexpression assays, and the physiological regulation of the localization
and activity of LKB1 in vivo therefore remains unclear.
The pseudokinase domain of the STRAD proteins binds
directly to the kinase domain of LKB1, enhancing its
catalytic activity over 100-fold. The molecular mechanism
by which STRAD activates LKB1 has not yet been elucidated, but it is possible that their interaction leads to a
conformational change that stabilizes LKB1 in an active
conformation (1). Interestingly, several single amino acid
substitution mutants of LKB1 isolated from patients have
lost the ability to interact with STRAD (20), pointing to
the importance of the STRAD-LKB1 interaction.
VII. THE METABOLIC AMPK MODULE
IS A DOWNSTREAM TARGET OF LKB1
B. MO25 Isoforms
Similar to STRAD, there are also two isoforms of
MO25, known as MO25␣ and MO25␤ (1, 18). Isoforms of
MO25 interact with the COOH-terminal Trp-Glu-Phe residues (“WEF motif”) of STRAD and stabilize the interaction between LKB1 and STRAD (18). Structural studies
have revealed that MO25␣ forms an extended ␣-helical
repeat rodlike structure, distantly related to the armadillo
repeat domain (82). At its COOH terminus, MO25␣ possesses a deep pocket that binds specifically to the WEF
motif of STRAD␣, and mutation of this pocket inhibited
the binding of MO25␣ to STRAD␣. Interestingly, a STRAD␣
mutant lacking the COOH-terminal WEF motif can still
form a complex with MO25␣ but only in the presence of
LKB1 (18, 20). This indicates that interaction of LKB1 with
STRAD␣ creates an additional binding site for MO25
within the complex, which may explain why MO25␣ stabilizes the binding of LKB1 to STRAD␣ (1). However, it
will be necessary to crystallize the entire LKB1-STRADMO25 complex to understand in full the molecular mechanism by which these proteins interact.
A. Physiological Regulation
Purification of the LKB1-STRAD-MO25 trimeric complex allowed for the characterization of the most prominent downstream LKB1 substrate identified so far in mammalian systems, the AMPK system (for review, see Ref.
41). AMPK has emerged as a master regulator of cellular
energy metabolism with dozens of downstream targets
C. Complex Formation and LKB1 Activation
Unlike most kinases, LKB1 is not activated by
phosphorylation of its activation loop by an upstream
kinase, but is instead activated upon binding to STRAD
(6). Complexes of LKB1-STRAD-MO25 can be isolated
from mammalian cells in which the three components
are present in a similar stoichiometry (18). LKB1 relocalizes from the nucleus to the cytosolic compartment
when coexpressed with MO25 and STRAD (6, 7, 22),
and the amount of STRAD␣ associated with LKB1 in
cells is significantly enhanced by the expression of
MO25␣. In addition, siRNA-mediated knockdown of
Physiol Rev • VOL
FIG. 3. AMPK mediates LKB1 signaling towards polarization and
energy metabolism. AMPK consists of an ␣-, ␤-, and ␥-subunit, representing the catalytic, regulatory, and AMP binding subunit, respectively.
Activation of AMPK is regulated by AMP binding to the ␥-subunit and
phosphorylation and dephosphorylation of the T-loop of the catalytic
subunit by upstream kinases and phosphatases, respectively (see text
for details). The fully active AMPK may then activate a multitude of
downstream targets involved in the regulation of protein translation and
cell growth, cell polarity, and cellular metabolism. Depending on context, several downstream pathways may be cooperatively activated, for
example, in the regulation of cellular polarity in response to changes in
cellular metabolism (53, 84). Government of cellular metabolic pathways and cell polarity may thus be broadly integrated under the control
of LKB1-AMPK family signaling.
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
unclear, as mutations that prevent STRAD from binding
ATP do not affect its ability to activate LKB1 or induce its
cytoplasmic localization (1). It has been speculated that
the STRAD proteins evolved from an active protein kinase
that perhaps once controlled LKB1 (1). However, there is
no evidence as yet to support this idea, as all vertebrate
and sea urchin homologs of STRAD also lack catalytic
residues (19). The NH2- and COOH-terminal domains are
not conserved between the STRAD isoforms, which may
account for functional differences between the two proteins with regard to complex formation (33). In this respect, it is interesting to note that recently a novel condition designated PMSE due to a homozygous deletion in
STRAD␣ (or LYK5) has been described which presents
with features of cortical thinning (91). Parents of patients
are hemizygous for the allele but apparently did not present
with any clinical abnormalities such as PJS polyps.
LKB1-AMPK FAMILY SIGNALING
Physiol Rev • VOL
AMPK activation, metabolic adaptation, and training endurance in conditional Lkb1 mouse models (see below).
B. LKB1 Is an AMPK Kinase
LKB1 was originally described as the upstream
AMPK kinase in response to studies investigating the
kinases responsible for activation of the yeast AMPK
ortholog snf1 (1). There are no clear orthologs of these
upstream kinases in the human genome, but one of the
protein kinases with a kinase domain closest to these
upstream kinases in yeast is human LKB1. Following this
observation, LKB1 was shown to efficiently phosphorylate AMPK at its regulatory T-loop residue Thr-172 in
biochemical assays (42, 104, 132). A requirement for
MO25 and STRAD in the process was shown by demonstrating that the ability of LKB1 to activate AMPK was
greatly enhanced when present in a trimeric complex.
Finally, it was shown that AMPK could not be activated by
known activators of AMPK such as phenformin or AICAR
in cell lines that lack LKB1 (42, 104, 132). AMPK activation was restored by stably expressing wild-type, but not
catalytically inactive, LKB1 (42). These cell lines do, however, display a basal level of AMPK activation (42, 104),
hinting at the possibility that LKB1 is not exclusively
responsible for AMPK activation. Other kinases such as
the calmodulin-dependent protein kinase kinases
(CaMKKs) and TAK-1 have since been shown to function as upstream kinases for the AMPK complex (41).
This might explain why loss of AMPK activation is not
a universal feature following conditional Lkb1 removal
in mice (12). LKB1 can phosphorylate both AMPK␣1
and AMPK␣2 with similar efficiency in vitro (42). Studies suggest that the LKB1 complex is constitutively
active in vivo, since treatments that activate AMPK
apparently do not change LKB1 catalytic activity (76,
94). Encompassing data described above showing that
AMP activates AMPK by allosteric activation as well as
by protecting the regulatory T-loop residue from Thr172 dephosphorylation suggest a model for AMPK activation wherein the phosphorylation state of Thr-172
depends on the relative rates of phosphorylation, catalyzed by upstream kinases such as LKB1 and CaMKKs,
and dephosphorylation, catalyzed by protein phosphatases that remain to be further characterized (98).
C. LKB1 Exhibits Isoform Specificity
Towards AMPK
Recent studies have painted a particularly interesting
picture with respect to the differential regulation of
AMPK isoforms by LKB1. In a conditional Lkb1 mouse
model, it was shown that in muscle lacking Lkb1, neither
contraction nor AICAR activates the AMPK␣2 isoform
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
(Fig. 3). Mammalian AMPK is a heterotrimeric complex
consisting of catalytic ␣-subunits (␣1 or ␣2), and regulatory ␤- (␤1 or ␤2), and ␥-subunits (either ␥1 or ␥2 or ␥3).
AMPK is activated by any cellular stressor that depletes
cellular ATP. Activation of AMPK results in downregulation of ATP-consuming pathways, while switching on
ATP-generating pathways (41). Activation of AMPK by
AMP occurs in two ways: AMP activates AMPK directly
via an allosteric mechanism by binding at the regulatory
␥-subunit, as well as indirectly by inhibiting dephosphorylation of a regulatory “T-loop” threonine residue within
the kinase domain at Thr-172 by protein phosphatases.
The combination of the two results in ⬎1,000-fold increase in kinase activity (112). The identity of the protein
phosphatase(s) acting on AMPK in vivo is unknown; however, the effect of AMP on dephosphorylation is substrate
mediated so AMP would be predicted to inhibit all phosphatases acting on AMPK. Moreover, a single point mutation in a domain of the regulatory ␥-subunit abolishes
both the inhibitory effect on dephosphorylation and the
direct allosteric activation (98). This point mutation occurs in an autosomal dominant form in the Wolff-Parkinson-White syndrome, a heart disorder characterized by
abnormal storage of glycogen in cardiac myocytes and
ventricular preexcitation leading to disturbances in cardiac conduction and rhythm abnormalities (41). AMPK
can be pharmacologically activated by AICAR (an AMP
mimic after intracellular conversion to AICAR-monophosphate), phenformin, and metformin. The latter is a widely
prescribed drug in the treatment of non-insulin-dependent
diabetes mellitus, which is thought to exert its blood
glucose-lowering effects through its ability to activate
AMPK in skeletal muscle and liver (139).
AMPK activation in vivo has both short-term and
long-term effects on energy metabolism. Acute effects
include increasing glucose transport and activating fatty
acid oxidation through phosphorylation of metabolic regulators. Repeated or prolonged activation of AMPK is
associated with increased expression of enzymes involved in glucose and lipid oxidation and the mitochondrial electron transport chain (121). These effects might
be established independently through either AMPK␣1 or
AMPK␣2 isoforms (see below). With regard to metabolic
adaptation through the induction of downstream metabolic targets, it was recently shown that AMPK activation
through prolonged AICAR administration induced a metabolic gene expression profile in untrained mice (88).
Remarkably, prolonged AICAR administration translated
into an increased running endurance in sedentary mice.
The role for LKB1 activation in these metabolic adaptations remains to be investigated, although LKB1 has been
shown to be responsible for some of the downstream
effects of AICAR, for example, on fatty acid oxidation
(117). It will therefore be of particular interest to determine the effect of prolonged AICAR administration on
785
786
JANSEN ET AL.
TABLE
2.
tion of AMPK in the liver is required for the ability of
metformin to lower blood glucose levels (105). Strikingly,
a recent study reported improved glucose tolerance and
insulin sensitivity after Lkb1 depletion restricted specifically to skeletal muscle (68). Thus loss of Lkb1 in either
liver or skeletal muscle leads to opposite effects on systemic glucose regulation. This might be explained by the
fact that hepatic deletion of Lkb1 leads to increases in
PGC1␣ (105), whereas skeletal muscle-specific Lkb1 depletion decreases PGC1␣ (68). PGC1␣ is thought to mediate transcription of key gluconeogenic enzymes (81).
Curiously, the Lkb1 mice that have an ⬃90% Lkb1 reduction in many tissues and ablation of Lkb1 activity in
skeletal muscle were reported to exhibit normal blood
glucose concentrations (95). It seems plausible that in this
case the beneficial effects of a reduction in skeletal muscle Lkb1 expression on glucose homeostasis are masked
by the detrimental effects of a reduction in hepatic Lkb1
expression. Collectively, these studies point to a role for
Lkb1 in metabolic regulation, predominantly through the
AMPK␣2 isoform.
D. mTOR Module
Particularly well-established downstream effects following AMPK activation are mediated through the TSC1/
TSC2-mTOR pathway, which elicit an inhibitory effect on
protein translation (41). AMPK phosphorylates and activates TSC1/TSC2, which suppresses the activity of the
small GTPase Rheb. Active GTP-bound Rheb normally
activates mTOR. However, TSC1/TSC2 activation
switches Rheb off through GDP conversion, thereby
inactivating mTOR. This leads to dephosphorylation of
the physiological mTOR effectors S6K and 4E-BP1, which
are involved in translation initiation. Recently, Raptor has
been identified as another downstream effector of AMPK
that may elicit a cellular energy stress program by inhibiting the mTOR module (40).
Previously, studies had suggested a role for the
mTOR module in LKB1 tumor suppression (1). This was
shown, at least partly, by the fact that mutations in TSC1
Overview AMPK␣1 and AMPK␣2 isoforms
Weight
Phosphorylation sites
Tissue distribution
Subcellular localization
Upstream activators
Activators
Inhibitors
Nuclear localization signal
Mouse model
AMPK␣1
AMPK␣2
62.8 kDa/550 amino acids
Thr-171 (LKB1, CAMKKs), Thr-258, Ser-458
Activity ubiquitous
Cytoplasmic
LKB1 complex, CAMKKs
AICAR, metformin, A-769662
Compound C*
No
Knock-out displays no phenotype (58)
62.3 kDa/552 amino acids
Thr-171 (LKB1, CAMKKs), Thr-258, Ser-491
Activity highest in skeletal muscle, liver, and heart
Nuclear, cytoplasmic
LKB1 complex, CAMKKs
AICAR, metformin, A-769662
Compound C*
Yes (residues 218–231) (114)
Insulin resistant, glucose intolerant (125)
Reference numbers are given in parentheses. * Note that compound C targets a broader range of kinases and is not a specific AMPK inhibitor (8).
Physiol Rev • VOL
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
(95). In this model, complete Lkb1 loss occurred only in
skeletal muscle through the expression of a Cre recombinase driven by the muscle creatine kinase promoter,
whereas other tissues such as testis, kidney, and lung
express ⬃10% of the normal level of Lkb1 due to expression of a hypomorphic allele. In these mice, Thr-172 phosphorylation of the AMPK␣1 isoform in response to prolonged overload increased similarly in normal and Lkb1depleted skeletal muscle; Thr-172 phosphorylation of
AMPK␣2 was not increased in control animals and completely absent in the Lkb1-depleted mice (81). These results suggest that in Lkb1-deficient muscle, the AMPK␣1
isoform is activated at the Thr-172 T-loop residue normally in response to chronic overload. Thr-172 phosphorylation of the AMPK␣2 isoform appears not to be involved
in chronic overload responses such as muscle hypertrophy (81). Therefore, it is becoming increasingly clear that
the AMPK␣1 and AMPK␣2 isoforms have distinct regulatory properties and functions in vivo and show discordant
regulation in response to metabolic stimuli (Table 2). It
has been proposed that the Lkb1-regulated AMPK␣2 isoform is involved in metabolic adaptations. Thus diminished AMPK␣2 isoform activation might be expected to
reveal itself predominantly in metabolic phenotypes such
as altered glucose regulation. This is in line with the fact
that knockout of the AMPK␣1 isoform in mice did not
result in a detectable metabolic phenotype, whereas
AMPK␣2 isoform knockout mice exhibit insulin resistance (58, 125). Note that these mice do not exhibit any of
the abnormalities associated with PJS, in particular PJS
polyposis, although tissue-specific redundancy cannot be
ruled out. Conversely, PJS patients have not been reported to demonstrate any metabolic or endocrine abnormalities.
Lkb1 depletion targeted to the liver has been shown
to result in reduced total hepatic AMPK activity and hyperglycemia (105). This was suggested to result from
uninhibited gluconeogenesis due to unabated activation
of the transcriptional coactivator Torc2. Moreover, hyperglycemia in mice lacking hepatic Lkb1 expression did not
respond to the blood glucose-lowering effects of metformin treatment, indicating that Lkb1-mediated activa-
LKB1-AMPK FAMILY SIGNALING
E. Clinical Applications
Following the demonstration that the tumor suppressor LKB1 is upstream of AMPK, it has been suggested that
the inhibition of cellular proliferation through AMPK activation may be used to prevent or even treat neoplastic
transformation (1). Retrospective epidemiological studies
have suggested that diabetics on metformin have a reduced likelihood of developing cancer (21, 35). Likewise,
AICAR treatment of cell lines in culture was shown to
inhibit proliferation (115). Recently, in an in vivo model,
activation of the AMPK pathway by metformin, phenformin, or A-769662 treatment was shown to significantly
delay tumor onset in Pten ⫹/⫺ mice also carrying a
hypomorphic mutation in Lkb1 (50). LKB1 was shown to
be required in vitro for activators of AMPK to inhibit
cellular proliferation in PTEN-deficient cell lines (50).
Although the concept of suppressing tumor development
by triggering a physiological signaling pathway that functions as a cellular energy-sensing checkpoint appears attractive, it needs to be remembered that AMPK as stated
previously displays a broad range of cellular targets (41).
Indeed, switching on a cellular energy checkpoint may
also trigger undesirable responses such as enhanced
angiogenesis. Furthermore, low-energy conditions may
make tumor cell populations less susceptible to standard treatments that depend on cellular turnover such
as radiation therapy or chemotherapy. Thus future
studies should investigate whether any of these pharmacological modalities such as metformin (already in
Physiol Rev • VOL
clinical use for the treatment of diabetes mellitus) can
translate into a significant benefit in the oncological
setting.
VIII. LKB1 AND CELLULAR POLARIZATION
Before the discovery of the energy-sensing AMPK
module as a downstream target, LKB1 had been most
prominently linked to the regulation of cellular polarization in model organisms. The allele encoding the Caenorhabditis elegans counterpart of the LKB1 tumor suppressor Par-4 was initially retrieved in screens designed to
pick mutants defective in asymmetric cell division of the
fertilized worm zygote. C. elegans Par-4 mutants show
symmetric cleavage patterns and absence of gut granules
showing that Par-4 is crucial for establishing cell polarity
during asymmetric cleavage patterns of blastomeres and
specification of the intestinal lineage (126). Likewise, the
Drosophila LKB1 counterpart dLkb1 was retrieved by
screening for mutants defective in anterior-posterior
oocyte axis formation (78). These studies remain some of
the strongest evidence linking LKB1 to cellular polarization and facets thereof. This role of LKB1 and recent
studies substantiating this role will be discussed next. We
will examine different canonical polarity model systems,
highlighting recent studies that have shown a surprising
convergence of LKB1-AMPK signaling on polarity establishment.
A. A Family of Par Proteins Regulates
Cellular Polarity
The acquisition and maintenance of cellular polarity
is a fundamental process both during development and
during adult tissue homeostasis. It has been found that a
core set of Par (for “partitioning defective”) proteins,
along with a limited number of other proteins such as
aPKC, are involved in a broad range of phenomena requiring proper cellular polarization (37). On the basis of work
performed initially in C. elegans employing genomic
screens looking for mutants demonstrating symmetric
first cleavage patterns (62), a family of six Par genes has
been cloned and characterized. In this family Par-1 and
Par-4 encode serine threonine kinases; Par-5 is a member
of the family of 14-3-3 proteins; Par-3 and Par-6 contain
PDZ domains, suggesting that they may act as part of a
signaling scaffold; and, finally, Par-2 displays a RING
finger domain which suggests it may act in a ubiquitination pathway. The latter protein is not conserved beyond
C. elegans, however. This core module of Par proteins
appears to be recursively recruited in several contexts of
cellular polarization, such as neurite extension, cellular
migration, and asymmetric cell division. A complex epistatic relationship for members of this family has been
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
and TSC2 are found in tuberous sclerosis, a disease that
features hamartomatous tumor formation in various organs. PJS polyps did show increased mTOR activation as
is evident from increased phosphorylation of its targets
S6K and 4E-BP1 (103). However, increased mTOR activation occurs in many instances of increased proliferation
of which the mature PJS polyp is but one example. Moreover, there is little clinical overlap between PJS and tuberous sclerosis, as stated above. A recent study investigating the effect of the mTOR inhibitor rapamycin in the
Lkb1 ⫹/⫺ mouse model found that rapamycin treatment
affects PJS polyposis by inhibiting the growth of established polyps, whilst smaller polyps were unaffected
(128). This might be explained by the fact that rapamycin
mainly inhibits proliferation, whereas it may not affect the
underlying mechanism of polyp formation. Moreover, activation of TSC2 and TORC1 is not affected by Lkb1
depletion in skeletal muscle (81). Evidence for a role of
mTOR in LKB1 signaling has remained scarce, and with
the demonstration of a role for AMPK in polarity regulation (see below), in addition to its well-described role in
energy metabolism, the latter pathway has since come
under increased scrutiny.
787
788
JANSEN ET AL.
FIG. 4. Par proteins establish cortical identity. From studies in
Drosophila, C. elegans, and mammalian cell culture systems, it has
become clear that various aspects of cell polarity are regulated by an
evolutionary conserved module of Par (“partitioning defective”) proteins, which is recursively iterated in various contexts of cellular polarization. Antagonistic phosphorylation events regulate the mutual exclusion through 14-3-3 association and segregation of Par proteins to different cortical domains. LKB1/Par-4 is a pivotal upstream regulator of
cellular polarization.
Physiol Rev • VOL
both modules and is therefore not asymmetrically localized (37). Whether a similar mode of operation applies for
LKB1, which also appears not to be asymmetrically localized, remains to be investigated.
Note that in some instances significant differences
between mechanisms of cellular polarization have been
observed between invertebrate and vertebrate model systems. It is clear, however, that similar to the mechanisms
deployed during asymmetric cell division in C. elegans
and Drosophila, progenitor cells within mammalian stem
cell niches, such as the interfollicular epidermis (70) and
the subventricular zone in the developing brain (for review, see Ref. 67), recruit Par family members to divide
asymmetrically. Studies that have begun to unravel the
signaling network downstream of LKB1 indicate that it
may signal to a host of intracellular targets depending on
context.
B. LKB1 in Epithelial Polarity
The formation of cellular sheets displaying apicobasal polarity is a defining attribute of several epithelial
tissues. Epithelia often perform a dual barrier function
with the environment, permitting the regulated uptake or
excretion of substances at its apical surface, whilst maintaining a closed, impenetrable surface through the formation of tight intercellular junctions. Notable examples of
this form of polarization include the intestinal epithelial
lining displaying apical brush borders and the ciliated
epithelium of the lung. Similar to the antagonistic modules described above, cellular polarization in mammalian
cell culture systems has been shown to proceed through
mechanisms whereby membrane domains are established
through mutual exclusion driven by phosphorylation, 143-3 association, and competitive binding (37, 52, 113).
Evidence on the domineering role played by LKB1 in
epithelial polarization was provided in a study showing
that single intestinal epithelial cells polarize in a cellautonomous fashion in response to the activation of LKB1
(7). Cellular polarization is accomplished through the induction of STRAD expression in these cells, which allows
for stabilization and activation of LKB1. Cells polarize
within hours after STRAD induction in the absence of
neighboring contacts. Polarized cells display several defining features of polarized intestinal epithelial cells, including reorganization of the actin cytoskeleton to form
an apical brush border, relocalization of junctional ZO-1
proteins surrounding the brush border, and sorting of
apical and basolateral membrane proteins to facilitate
directed endosomal transportation (7). These observations suggest that LKB1 acts upstream of the other Par
members in mammalian epithelial cells, similar to genetic
data retrieved from Drosophila and C. elegans. For instance, the observation that putative junctional com-
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
revealed by a host of functional studies (37). LKB1/Par-4
may reside at the top of a cascade regulating the asymmetric segregation of Par family proteins, since LKB1/
Par-4 mutations are epistatic to all other Par family mutants (60). Combining data from several model systems
suggests a model wherein antagonistic activities of the
different Par modules regulate the identity of different
cellular membrane domains (Fig. 4) (37). In this scheme,
Par-1/MARK may phosphorylate Par-3 on two distinct
serine residues generating binding sites for Par-5/14-3-3,
which inactivates Par-3 and prevents it from invading the
basolateral domain by inhibiting Par-3/Par-6/aPKC complex formation (13). Conversely, aPKC may phosphorylate Par-1/MARK, which allows for 14-3-3 binding and
inhibits its function at the apical domain. Since LKB1 sits
at the top of this cascade, its function may be to “tilt the
balance” between these opposing kinase modules in different membrane domains or in different contexts. Indeed, it has been shown that LKB1 can phosphorylate
Par-1, resulting in the activation of this kinase (see below)
(76, 110). 14-3-3 is required for reciprocal exclusion of
LKB1-AMPK FAMILY SIGNALING
Physiol Rev • VOL
titude of downstream LKB1 substrates now implicates
LKB1 signaling in an array of cellular processes ranging
from metabolic control to cellular polarization. As observed for AMPK, STRAD and MO25 are essential for
LKB1 to phosphorylate AMPK-related enzymes. LKB1
does not activate all kinases in this branch, and specificity
appears to be partly mediated by a preference for a
leucine at position ⫺2 relative to the T-loop. Among the
kinases in this family of AMP kinases are the four mammalian homologs of the Par-1 kinase, MARK1– 4 (for microtubule affinity-regulating kinase 1– 4) and the BRSK1/
SAD-A, BRSK2/SAD-B kinases. The MARK kinases will be
discussed next, and the BRSK1/SAD-A, BRSK2/SAD-B kinases will be discussed more extensively thereafter. For a
more comprehensive review on the other downstream
AMPK family substrates of LKB1, the interested reader is
referred to Reference 1.
Extensive evidence exists to support a conserved
function of the MARK family members in regulating aspects of mammalian epithelial polarization. The four
closely related MARK kinases are the mammalian counterparts of the Drosophila and C. elegans Par-1 kinase,
which function cooperatively with LKB1/Par-4 in the establishment of cellular polarity in these model organisms
(14, 78). Previous studies indicated that the four members
of this AMPK-related kinase subfamily play similar roles
in regulating cell polarity at a mammalian level (15, 25, 34,
113). The demonstration that MARK isoforms are phosphorylated and activated by LKB1 (76, 110) suggested that
LKB1/Par-4 regulates cell polarity by activating MARK/
Par-1 isoforms. These phosphorylation events may lead to
the establishment of distinct cellular membrane domains
through mutual exclusion with the Par-3/6 module according to the scheme outlined above. Proteomic analyses suggest that this network is conserved at the mammalian level (22).
In mammalian cells, MARK isoforms are a major
determinant of the organization of the microtubule skeleton through the phosphorylation of microtubule-associated proteins (34). This may cause detachment of the
latter from microtubules and induce microtubule destabilization (15). Thus MARK activation may locally regulate
microtubule density and mediate critical aspects of cellular polarization, such as centrosome repositioning. In an
elegant study by Taketo and co-workers (69), LKB1 depletion was found to affect microtubule dynamic instability by suppressing microtubule polymerization through
activation of the AMPK family member MARK2 (69). This
work accords with earlier data from the Drosophila system, wherein germline dLkb1 clones show loss of the
polarized microtubule network in the Drosophila oocyte
(78). Two somatic point mutations of MARK3 have recently been reported in sporadic colorectal cancer (89),
suggesting that MARK3 may be causally involved in tumorigenesis. The genes encoding these MARK isoforms
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
plexes form in the absence of neighboring cells indicates
that LKB1 can regulate Par-3 and Par-6 by establishing the
structures that recruit this complex. Obviously, this does
not exclude other pathways that emanate from LKB1 to
regulate the Par-3 complex, or vice versa. The exact role
of 14-3-3/Par-5 in this context remains unclear; recent
work indicates that 14-3-3␨ can interact with phosphorylated MARK family members, thereby blocking plasma
membrane localization of the latter (38).
Studies examining epithelial polarization in the
MDCK cell culture model system have provided evidence
for a remarkable role of AMPK activation in polarity
establishment, in addition to its role in energy metabolism. In mammalian MDCK cell culture models, activation
of AMPK appears to be required for repolarization of
MDCK cells in response to changes in extracellular calcium (“calcium switch” model) (136, 138). Experiments
using kinase-dead LKB1 showed that activation of AMPK
during cellular repolarization was dependent on LKB1.
This work did not investigate whether expression of kinase-dead LKB1 translated into polarization defects (136).
Note that, in contrast to studies demonstrating a differential regulation of the AMPK␣1 and AMPK␣2 isoforms in
skeletal muscle by Lkb1, these studies have not examined
whether in epithelial cells the various AMPK␣ isoforms
are differently involved. Important support for these observations in mammalian cell culture models has now
come from two studies examining the phenotype of the
Drosophila AMPK null mutant. Mutants were embryonic
lethal and showed extensive defects in cellular polarization as shown by the mislocalization of specific membrane markers (72, 84). Thus AMPK signaling in polarity
establishment appears to be evolutionary conserved. Active mutants of myosin regulatory light chain (MRLC)
rescued many of the defects associated with AMPK loss,
suggesting that MRLC is principally responsible for executing signaling downstream of AMPK in this context
(72). This is a remarkable finding as AMPK displays an
array of downstream targets. This would suggest a conserved role for actomyosin contractility in cellular polarization (28). With regard to the regulation of MRLC in
mammalian systems, activation of AMPK has been shown
to correlate with MRLC inhibition rather than activation
in vascular smooth muscle cells (49), and deletion of Lkb1
in pancreas did not affect MRLC phosphorylation status
(48). Further research will therefore be required to resolve these issues.
Following the identification of LKB1 as the upstream
kinase of AMPK, it was quickly realized that LKB1 might
activate other kinases akin to AMPK. Indeed, LKB1 was
shown to function upstream of 12 additional kinases that
fall on the same proteomic branch as AMPK by phylogenetic analyses (1, 76). These kinases are all activated by
LKB1 through phosphorylation of their “T-loop” threonine
residue, which is required for full activity (76). This mul-
789
790
JANSEN ET AL.
C. LKB1 in Neuronal Polarity
Recent studies have examined the role of LKB1 in
neuronal polarization and found additional downstream
substrates. In the developing neocortex, neural progenitors migrate away from their birthplace within the ventricular zone to differentiate and establish synaptic connections in the cortical plate by assuming proper axonal/
dendritic polarity. Neuronal migration and differentiation
are dependent on a coordination of events, involving neurite outgrowth and centrosome relocation. Neurite extension is a classic model for research on mechanisms of
cellular polarization, and a pivotal role for Par family
members has been demonstrated in this system as well
(37). Loss of a conditional Lkb1 allele in telencephalic
neuronal progenitors impairs corticofugal axon extension, resulting in cortical thinning and agenesis of the
corpus callosum in mice (12). This phenotype was traced
specifically to a defect in activation of the Brsk/Sad kinases, which appear to be responsible for cytoskeletal
reorganization through tau phosphorylation (12). Previously, Sad-A/Sad-B combined knockout mice had shown
a similar phenotype of reduced axon outgrowth and mislocalization of axon and dendritic markers (65). These
SAD kinases are expressed mainly in the brain and are
members of the AMPK family of downstream LKB1 effectors, which are phosphorylated on the T-loop residue in
the activation domain like the Par-1/MARK subfamily of
kinases (see above). No changes in AMPK phosphorylation in Lkb1-deficient cortex were noted in this study (4,
Physiol Rev • VOL
111). Additionally, extracellular cues such as brain-derived neurotrophic factor (BDNF) were shown to induce
axon formation in vitro by locally activating LKB1 phosphorylation on serine-431 (130). A third study took a
similar approach using in vivo electroporation of RNAi
constructs targeting Lkb1 transcripts into neonatal mouse
cortex to investigate the effects of Lkb1 depletion. Here,
it was similarly found that neuronal migration and axonal
polarity were impaired, which were linked to a malpositioning of the centrosome in migrating and differentiating
neurons. LKB1 has similarly been found to affect polarized migration in response to wound healing in cell culture models of migrating epithelial cells, which may be
regulated through the small GTPase Cdc42 (137). Polarized migration involves major cytoskeletal reorganization
events, involving Golgi and centrosome repositioning and
lamellipodia formation. Localized LKB1 activation and
subsequent microtubule reorganization, through for example tau phosphorylation (69), may depend in vivo on
neurite outgrowth-promoting cues such as BDNF (12). A
role for AMPK in polarized migration is conceivable regarding its instructive role in the establishment in epithelial polarity across systems, although evidence for such a
role remains scarce at this point. Collectively, current
data make it clear that Lkb1 deficiency in cortical neurons
leads to a cell-autonomous defect in axon formation and
appear to reaffirm the critical role of LKB1 in polarity
regulation in vivo. A key challenge of future research will
be to identify the substrates BRSK/SAD and MARK kinases phosphorylate to regulate cell polarity in vivo.
Clearly, the role of LKB1 during cellular polarization is
not limited to epithelial systems.
D. Role of LKB1 in Asymmetric Cell Division
The work discussed so far examined a role for LKB1AMPK signaling in cell polarization during interphase.
However, a large body of work has documented a role for
LKB1 during mitosis as well. Cell division involves many
aspects similar to interphase cellular polarization, and
similar signaling pathways will therefore be reiterated. In
fact, the original Par-4 allele, which allowed for the cloning of the C. elegans LKB1 homolog, was retrieved in
screens examining asymmetric cell division long before
downstream LKB1 effectors such as the AMPK family
members were functionally characterized (62). The Par
genes that were retrieved in C. elegans are required for
two related aspects of cell polarization during asymmetric
cell division: the asymmetric placement of the mitotic
spindle, which results in unequal division, and the asymmetric positioning of factors determining cell cycle timing
and cell fate distinction between daughter cells, the socalled cytoplasmic determinants (Fig. 5) (14). Interestingly, Par-4 mutants do show asymmetric first cleavage
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
would represent potential secondary PJS loci; however,
mutation analysis did not reveal germline mutations in
PJS patients not previously linked to LKB1 mutations
(29). In addition to LKB1, another kinase named TAO1
was shown to activate MARK2 by T-loop phosphorylation
(119). This suggests that LKB1 may not exclusively mediate activation of downstream substrates in the MARK
family, as has similarly been shown for CAMKK and TAK1
with regard to AMPK. The contribution of the MARK
family members to the establishment of cellular polarization remains under investigation. MARK knockout mice
have been established, but these did not reveal obvious
defects in cellular polarization, although redundancy
among the four family members may have masked functional defects in this respect. Strikingly, MARK2 (also
called Par-1b) knockout mice have recently been found to
be hypoinsulinemic and thus display a metabolic phenotype (51). These mice had increased glucose tolerance but
were normoglycemic under fasting conditions. This indicates that MARK family members may function in the
regulation of metabolic homeostasis in addition to cellular polarization, echoing results obtained for the AMPK
module.
LKB1-AMPK FAMILY SIGNALING
patterns, albeit with larger variation (86, 126). Divisions in
the two-cell Par-4 embryo occur with near-perfect synchronicity, which is in stark contrast to the wild-type
embryo, wherein divisions are strictly asynchronous.
Thus Par-4 does act during the first cell cycle, as mutants
do not segregate cytoplasmic determinants at this time
even though the actual cleavage pattern is asymmetric.
Par-4 therefore demonstrates an uncoupling of spindle
placement and determinant segregation during asymmetric cell division in the C. elegans zygote. In Drosophila,
oocyte polarization occurs before, not during, fertilization
as in C. elegans, yet with the exception of Par-2, which is
not conserved in Drosophila, the same set of proteins
appears to be involved (14). The Drosophila LKB1 ortholog dLkb1 has been found to regulate the establishment of oocyte polarization, and mutation of dLkb1 is
embryonic lethal (78). The Drosophila Par-1/MARK homolog has also been found to cooperate with dLkb1 in this
process (78, 118). Moreover, similar to Par-4 in C. elegans, a
recent forward genetic screen in the Drosophila model has
uncovered a role for dLkb1 during neuroblast asymmetric
cell division (16). Neuroblast cell division represents one
Physiol Rev • VOL
of the best-characterized model systems for asymmetric
stem cell division (for a review, see Ref. 67), and key
players recovered in this system (such as Lgl and Pins)
have been found to play a conserved role during asymmetric cell division in higher organisms as well (66, 70).
Asymmetric neuroblast division is characterized by the
asymmetric inheritance of segregating determinants and
the coupling of mitotic spindle orientation to determinant
localization. Mutations in regulators of either process
have been shown to result in a hyperproliferation phenotype due to occasional neuroblast divisions in mutant
brains undergoing ectopic self-renewal through symmetric cell division (71). Mutation of the Drosophila LKB1
homolog resulted in disruption of many aspects of asymmetric larval neuroblast division, such as asymmetric microtubule length and centrosome size, and disruption of
asymmetric distribution of lineage determinants. Neuroblasts mutant for dLkb1 were shown to undergo occasional ectopic symmetric divisions, and mutant brains
demonstrate a phenotype of hyperproliferation, accordingly. The Drosophila STRAD and MO25 homologs were
also shown to be involved in neuroblast division in a
second study (133).
At this point, direct evidence for a role of LKB1 in
mammalian asymmetric cell division has not been put
forth. However, based on loss of function studies in invertebrate model systems, LKB1 may have a role in directing asymmetric cell division in higher organisms as
well. Lineage determination in several adult mammalian
stem cell compartments also proceeds via asymmetric
stem cell division occurring in stem or progenitor cells
(67). We have recently shown that normal unaffected
intestinal epithelium in PJS patients shows an expanded
progenitor compartment (30). Previous studies have indicated that the mammalian intestinal epithelium is maintained through asymmetric progenitor cell divisions in the
intestinal stem cell compartment (10). In addition, progenitor cells undergo symmetric divisions during normal
homeostasis, which may dictate the expansion or loss of
intestinal progenitor cell lineages within the stem cell
compartment (Fig. 6) (73). An expanded progenitor compartment as observed in normal unaffected Peutz-Jeghers
epithelium may be due to occasional ectopic symmetric
progenitor cell divisions, in accordance with the consequences of LKB1 mutation in invertebrate model organisms. Stochastic bursts of epithelial proliferative activity
may form mucosal outpocketings upon which mechanical
forces can act. Potentially these lesions wax and wane
over time (smaller lesions may be extruded or auto-amputated) and fail to come to clinical attention (123). A
“tug-of-war” between small mucosal elevations and mechanical forces may precipitate the mature irregular, cauliflower-like PJS polyp (Fig. 1), which is typically located
at sites of greatest mechanical traction, such as the pylo-
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
FIG. 5. Phenotypic summary of Par mutant embryos in C. elegans.
The fertilized C. elegans zygote divides asymmetrically to generate two
daughter cells of dissimilar developmental potential. Par mutations affect either the segregation of cytoplasmic determinants (“P granules”),
the eccentric placement of the spindle, or both of these processes.
791
792
JANSEN ET AL.
normal unaffected epithelia (Fig. 6). In the colon, time to
gatekeeper mutation would therefore be decreased in PJS
patients at normal background mutation rates. In this
scenario, colorectal tumor progression in PJS patients
would be expected to adhere to a conventional adenomacarcinoma progression model; features of the mature PJS
polyp such as smooth muscle proliferation may be acquired coincidentally later during adenoma development
(see sect. IIB on clinical characteristics). Although evidence for this notion remains scarce at the moment, it has
been suggested that PJS patients indeed develop adenomatous polyps at an increased rate compared with the
general population (80). In contrast, transgenic Lkb1
mouse models have not been shown to develop adenomatous polyps so far. This may relate to the fact that these
mice succumb to intestinal blockage at a time before
adenomatous changes in the murine colon has had time to
evolve. Thus, with regard to tumor progression in the
FIG. 6. Lineage competition in the intestinal stem cell crypt.
A: model depicting asymmetric progenitor cell division. Stem cell lineages may either be driven towards extinction or expansion by symmetric progenitor cell divisions resulting in two daughter cells committed to
differentiate or two daughter cells retaining progenitor cell characteristics, respectively. Asymmetric progenitor cell divisions leave the progenitor cell population unchanged. B: clonal evolution in stem cell
compartments. Depicted is a hypothetical progenitor compartment (a
“niche”), which contains two progenitor lineages at baseline. One of
these progenitor cell lineages has sustained a neutral polymorphic
marker mutation (for example, loss of O-acetyltransferase activity).
During progenitor cell turnover, lineages are continuously lost to follow-up due to random loss of lineages with replacement by others. The
rate of lineage loss is governed by the symmetric division rate as shown
in A. New marker alleles arise constantly at a normal background
mutation rate. In this example, one progenitor lineage carrying an oncogenic mutation (for example, p53 mutation or K-ras activation) arises.
As a result of random genetic drift, this lineage may either attain
dominance (bottom) or be lost to follow-up (top). Thus progenitor cell
populations in a niche and their pool of accumulated mutations continuously change during life as a result of lineage competition. C: normal
unaffected Peutz-Jeghers epithelium demonstrates an expanded progenitor compartment. D and E: an expansion of the progenitor pool represents a cancer-prone state. D depicts a hypothetical series of wild-type
adult colon crypts, whereas E represents a series of Peutz-Jeghers colon
crypts. A greater number of selectable variants is retained over a longer
period of time among the pool of progenitors in Peutz-Jeghers epithelium in E due to protracted clonal evolution of the niche as a result of an
expanded progenitor cell pool. An increase in the pool of selectable
variants accelerates the time to neoplastic transformation, since it increases the likelihood that a fortuitous combination of oncogenic mutations (for example, a biallelic APC mutation) occurs. Mutations (neutral and oncogenic) are sustained under both scenarios at a similar
normal background mutation rate. However, variations in clonal evolution rate allow mutations to accumulate at different frequencies even
though mutations arise at a similar frequency throughout life. This
lottery-like process is played out at steady-state in each of the ⫾15
million crypts that constitute the adult human colon. In the example
shown, one lineage carrying an oncogenic mutation has drifted to dominance in the crypt (right). Note that this process is entirely obscure on
routine histopathology, as we have no markers of prospectively tracing
individual stem cell lineages. In this scenario, LKB1 hemizygosity exerts
its oncogenic effects before visible neoplastic transformation, and
Peutz-Jeghers syndrome therefore models an accelerated pretumorigenesis.
Physiol Rev • VOL
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
rus (92) or rectum (129), and displays histopathological
features of mucosal prolapse accordingly (54).
Malignant tumor progression is characterized by an
accumulation of genetic lesions through recurrent waves
of clonal expansion and the tractable acquisition of dysplastic features. In the colon, this multistep progression
model follows a sequence of histopathological steps from
normal epithelium to small adenoma, larger adenoma
with high-grade dysplasia, and eventually invasive carcinoma. It remains unresolved whether PJS polyps carry an
inherent risk for neoplastic transformation (54). As stated
previously, it is clear that polyp development and neoplastic transformation must both be accounted for by the
same genetic mechanism, regardless of whether PJS polyps per se are ultimately deemed to lack an inherent risk
of transformation. With regard to the cancer-prone state
in PJS patients, an expanded progenitor pool is predicted
to accelerate the time to neoplastic transformation in
LKB1-AMPK FAMILY SIGNALING
Physiol Rev • VOL
sion of a polar body in a primary oocyte represent a
particularly remarkable example of asymmetric cell division in the human system, both in terms of asymmetric
cell size and cell fate. One interpretation consistent with
current data from invertebrate model organisms is that
these neoplasms develop due to disrupted asymmetric
cell division occurring in meiosis I in female PJS patients.
IX. CONCLUSION
Taken together, current data favor a role for LKB1 in
the regulation of cellular polarization through activation
of a diverse set of targets such as the AMPK, MARK, or
SAD kinases (Fig. 7). It is worth bearing in mind though
that Lkb1-deficient mice, remarkably, do not display polarity-related phenotypes during early embryogenesis
(134). This suggests that, depending on context, redundant mechanisms may be involved in regulation of polarity establishment. The role of AMPK signaling remains to
be more thoroughly characterized. Evidence from a mammalian in vivo model regarding defects in cellular polarization has not been reported, even though data from
invertebrate systems and in vitro cell culture models suggest a dominant role for AMPK activation. LKB1 appears
to regulate the AMPK␣2 isoform, the activity of which is
most prominent in liver, heart, and skeletal muscle
(Table 2) (4, 111). Moreover, AMPK␣2 knockout mice
present a metabolic phenotype, unlike the phenotype of
the Lkb1 ⫹/⫺ mice. However, rescue by the AMPK␣1
isoform cannot be excluded in this case. Thus definitive
FIG. 7. LKB1 regulates cellular polarization in various contexts.
Polarization is an instrumental aspect of a range of cellular processes,
and many studies now suggest a key role for LKB1/Par-4. Examples of
polarized processes are asymmetric cell division, apical and basolateral
domain determination in epithelial cells, axon specification in neurons,
and directed migration of fibroblasts/mesenchymal cells.
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
intestinal tract, PJS patients develop PJS polyps that may
serve as a pointer to the cancer-prone condition and
adenomatous polyps that are the focus of neoplastic
change.
Since clonal evolution of stem cell compartments is
obscure on routine histopathology, LKB1 hemizygosity in
PJS patients (or in the sporadic setting) is expected to
exert its oncogenic effect during “pretumor progression”
before visible neoplastic transformation (63). Pretumor
progression refers to tumor progression in the absence of
phenotypic changes and specifies that stem cells accumulate mutations stochastically from birth in phenotypically
normal epithelia; pretumor progression therefore precedes gatekeeper mutation. A protracted clonal evolution
scenario resulting from an expanded progenitor compartment accelerates pretumor progression by expanding the
pool of selectable variants (Fig. 6). This oncogenic effect
of LKB1 hemizygosity on stem cell lineage turnover and
the tempo of accumulation of genetic changes in phenotypically unaffected progenitor compartments would similarly pertain to other stem cell niches in the mammalian
system, for example, progenitor cell compartments in the
epithelial parenchyma of the lung, breast, or exocrine
pancreas. Inactivation of the wild-type LKB1 allele may
afford selective advantages during visible clonal expansion in terms of, for example, an enhanced migratory
potential. Formally testing this concept on the involvement of LKB1 in the regulation of asymmetric stem cell
division would require lineage tracing experiments wherein
multiple lineages can be traced competing over time in
individual crypts in normal and PJS epithelia. Although
the mechanism underlying the expanded progenitor compartment remains unclear (whether this may be due either
to a cell-autonomous defect in asymmetric stem cell division or due to faulty epithelial-mesenchymal cross-talk), it
is clear that LKB1 hemizygosity is not neutral in adult
mammalian intestinal epithelial homeostasis. Future studies will address the role of LKB1 in mammalian epithelial
stem cell turnover. Demonstration of an expanded stem
cell pool in normal PJS epithelium would directly implicate a cancer-prone state at normal background mutation
rates.
Recent work from our own laboratory has now definitively shown that crypt base columnar cells constitute
a stem cell pool of the adult mammalian small intestine
and colon (11). Lgr5 represents the first adult mammalian
stem cell marker, and it should facilitate research into
possible pathways regulating stem cell turnover affected
by LKB1. Although evidence exists for a role of LKB1 in
asymmetric cell division in invertebrate model organisms,
so far this has not been reported in the mammalian setting. It is noteworthy in this respect that female PJS
patients almost universally display characteristic benign
germ cell-derived tumors of the ovaries called SCTATs, as
outlined above (135). Meiotic cell division and the extru-
793
794
JANSEN ET AL.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
ACKNOWLEDGMENTS
M. Jansen and J. P. ten Klooster both contributed equally to
the manuscript.
Address for reprint requests and other correspondence: H.
Clevers, Hubrecht Institute, Developmental Biology and Stem
Cell Research, Uppsalalaan 8, 3584 CT Utrecht, The Netherlands
(e-mail: [email protected]).
17.
18.
19.
REFERENCES
20.
1. Alessi DR, Sakamoto K, Bayascas JR. LKB1-dependent signaling
pathways. Annu Rev Biochem 75: 137–163, 2006.
2. Alhopuro P, Katajisto P, Lehtonen R, Ylisaukko-Oja SK,
Naatsaari L, Karhu A, Westerman AM, Wilson JH, de Rooij
FW, Vogel T, Moeslein G, Tomlinson IP, Aaltonen LA, Makela
TP, Launonen V. Mutation analysis of three genes encoding novel
Physiol Rev • VOL
21.
LKB1-interacting proteins, BRG1, STRADalpha, and MO25alpha, in
Peutz-Jeghers syndrome. Br J Cancer 92: 1126 –1129, 2005.
Alhopuro P, Phichith D, Tuupanen S, Sammalkorpi H, Nybondas M, Saharinen J, Robinson JP, Yang Z, Chen LQ, Orntoft T,
Mecklin JP, Jarvinen H, Eng C, Moeslein G, Shibata D,
Houlston RS, Lucassen A, Tomlinson IP, Launonen V, Ristimaki A, Arango D, Karhu A, Sweeney HL, Aaltonen LA.
Unregulated smooth-muscle myosin in human intestinal neoplasia. Proc Natl Acad Sci USA 105: 5513–5518, 2008.
Arad M, Seidman CE, Seidman JG. AMP-activated protein kinase in the heart: role during health and disease. Circ Res 100:
474 – 488, 2007.
Aretz S, Stienen D, Uhlhaas S, Loff S, Back W, Pagenstecher
C, McLeod DR, Graham GE, Mangold E, Santer R, Propping P,
Friedl W. High proportion of large genomic STK11 deletions in
Peutz-Jeghers syndrome. Hum Mutat 26: 513–519, 2005.
Baas AF, Boudeau J, Sapkota GP, Smit L, Medema R, Morrice
NA, Alessi DR, Clevers HC. Activation of the tumour suppressor
kinase LKB1 by the STE20-like pseudokinase STRAD. EMBO J 22:
3062–3072, 2003.
Baas AF, Kuipers J, van der Wel NN, Batlle E, Koerten HK,
Peters PJ, Clevers HC. Complete polarization of single intestinal
epithelial cells upon activation of LKB1 by STRAD. Cell 116: 457–
466, 2004.
Bain J, Plater L, Elliott M, Shpiro N, Hastie CJ, McLauchlan
H, Klevernic I, Arthur JS, Alessi DR, Cohen P. The selectivity
of protein kinase inhibitors: a further update. Biochem J 408:
297–315, 2007.
Bardeesy N, Sinha M, Hezel AF, Signoretti S, Hathaway NA,
Sharpless NE, Loda M, Carrasco DR, DePinho RA. Loss of the
Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 419: 162–167, 2002.
Barker N, van de Wetering M, Clevers H. The intestinal stem
cell. Genes Dev 22: 1856 –1864, 2008.
Barker N, van Es JH, Kuipers J, Kujala P, van den Born M,
Cozijnsen M, Haegebarth A, Korving J, Begthel H, Peters PJ,
Clevers H. Identification of stem cells in small intestine and colon
by marker gene Lgr5. Nature 449: 1003–1007, 2007.
Barnes AP, Lilley BN, Pan YA, Plummer LJ, Powell AW,
Raines AN, Sanes JR, Polleux F. LKB1 and SAD kinases define
a pathway required for the polarization of cortical neurons. Cell
129: 549 –563, 2007.
Benton R, St Johnston D. Drosophila PAR-1 and 14-3-3 inhibit
Bazooka/PAR-3 to establish complementary cortical domains in
polarized cells. Cell 115: 691–704, 2003.
Betschinger J, Knoblich JA. Dare to be different: asymmetric
cell division in Drosophila, C. elegans, and vertebrates. Curr Biol
14: R674 – 685, 2004.
Biernat J, Wu YZ, Timm T, Zheng-Fischhofer Q, Mandelkow
E, Meijer L, Mandelkow EM. Protein kinase MARK/PAR-1 is
required for neurite outgrowth and establishment of neuronal polarity. Mol Biol Cell 13: 4013– 4028, 2002.
Bonaccorsi S, Mottier V, Giansanti MG, Bolkan BJ, Williams
B, Goldberg ML, Gatti M. The Drosophila Lkb1 kinase is required
for spindle formation and asymmetric neuroblast division. Development 134: 2183–2193, 2007.
Bosman FT. The hamartoma-adenoma-carcinoma sequence.
J Pathol 188: 1–2, 1999.
Boudeau J, Baas AF, Deak M, Morrice NA, Kieloch A, Schutkowski M, Prescott AR, Clevers HC, Alessi DR. MO25alpha/
beta interact with STRADalpha/beta enhancing their ability to bind,
activate and localize LKB1 in the cytoplasm. EMBO J 22: 5102–
5114, 2003.
Boudeau J, Miranda-Saavedra D, Barton GJ, Alessi DR.
Emerging roles of pseudokinases. Trends Cell Biol 16: 443– 452,
2006.
Boudeau J, Scott JW, Resta N, Deak M, Kieloch A, Komander
D, Hardie DG, Prescott AR, van Aalten DM, Alessi DR. Analysis of the LKB1-STRAD-MO25 complex. J Cell Sci 117: 6365– 6375,
2004.
Bowker SL, Majumdar SR, Veugelers P, Johnson JA. Increased
cancer-related mortality for patients with type 2 diabetes who use
sulfonylureas or insulin. Diabetes Care 29: 254 –258, 2006.
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
demonstration for a role of AMPK signaling in mammalian cellular polarization and PJS pathogenesis awaits a
conditional mouse model that will allow targeting both
AMPK isoforms simultaneously.
The mechanism responsible for the cancer-prone
state in PJS patients remains unclear. The majority of
studies investigating LKB1 function are performed under
LKB1 null conditions. Although this has revealed valuable
functional information on the consequences of LKB1 loss,
it needs to be underscored that LKB1 hemizygosity is not
silent, both with respect to the suppression of polyposis
and the suppression of malignant transformation. Thus
issues of gene dosage and tissue heterogeneity will need
to be taken into account when considering the target(s) of
LKB1. It remains unproven that PJS polyps carry an increased potential for malignant transformation compared
with the surrounding unaffected epithelium. The molecular mechanism for polyp development must, however,
underlie malignant transformation in PJS patients as well.
Data from invertebrate model organisms have indicated
an important role for LKB1 in the regulation of asymmetric stem cell division. Ectopic symmetric progenitor
cell divisions may result in an expanded progenitor cell
compartment as observed in unaffected PJS epithelium.
This may independently mediate both PJS polyp development and the cancer-prone state in PJS through one
molecular mechanism.
The demonstration that the LKB1 locus is affected in
tumors from a significant number of sporadic lung cancer
patients has increased interest in this enigmatic tumor
suppressor protein. Clearly, the importance of research in
patients afflicted by rare tumor predisposition syndromes
such as PJS is validated by these observations. Current
work implicating LKB1 in metabolic adaptation and training endurance further increases interest in the potential
for drugs targeting LKB1 signaling. If the pace of research
over recent years is a yardstick for future progress, then
we can hope to see some exciting developments in the
LKB1 field soon.
LKB1-AMPK FAMILY SIGNALING
Physiol Rev • VOL
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
kinases in the AMP-activated protein kinase cascade. J Biol 2: 28,
2003.
Hearle N, Schumacher V, Menko FH, Olschwang S, Boardman
LA, Gille JJ, Keller JJ, Westerman AM, Scott RJ, Lim W,
Trimbath JD, Giardiello FM, Gruber SB, Offerhaus GJ, de
Rooij FW, Wilson JH, Hansmann A, Moslein G, Royer-Pokora
B, Vogel T, Phillips RK, Spigelman AD, Houlston RS. Frequency and spectrum of cancers in the Peutz-Jeghers syndrome.
Clin Cancer Res 12: 3209 –3215, 2006.
Hearle NC, Rudd MF, Lim W, Murday V, Lim AG, Phillips RK,
Lee PW, O’Donohue J, Morrison PJ, Norman A, Hodgson SV,
Lucassen A, Houlston RS. Exonic STK11 deletions are not a rare
cause of Peutz-Jeghers syndrome. J Med Genet 43: e15, 2006.
Hemminki A, Markie D, Tomlinson I, Avizienyte E, Roth S,
Loukola A, Bignell G, Warren W, Aminoff M, Hoglund P,
Jarvinen H, Kristo P, Pelin K, Ridanpaa M, Salovaara R, Toro
T, Bodmer W, Olschwang S, Olsen AS, Stratton MR, de la
Chapelle A, Aaltonen LA. A serine/threonine kinase gene defective in Peutz-Jeghers syndrome. Nature 391: 184 –187, 1998.
Hemminki A, Tomlinson I, Markie D, Jarvinen H, Sistonen P,
Bjorkqvist AM, Knuutila S, Salovaara R, Bodmer W, Shibata
D, de la Chapelle A, Aaltonen LA. Localization of a susceptibility
locus for Peutz-Jeghers syndrome to 19p using comparative
genomic hybridization and targeted linkage analysis. Nature Genet
15: 87–90, 1997.
Hernan I, Roig I, Martin B, Gamundi MJ, Martinez-Gimeno M,
Carballo M. De novo germline mutation in the serine-threonine
kinase STK11/LKB1 gene associated with Peutz-Jeghers syndrome.
Clin Genet 66: 58 – 62, 2004.
Hezel AF, Gurumurthy S, Granot Z, Swisa A, Chu GC, Bailey
G, Dor Y, Bardeesy N, Depinho RA. Pancreatic lkb1 deletion
leads to acinar polarity defects and cystic neoplasms. Mol Cell Biol
28: 2414 –2425, 2008.
Horman S, Morel N, Vertommen D, Hussain N, Neumann D,
Beauloye C, El Najjar N, Forcet C, Viollet B, Walsh MP, Hue
L, Rider MH. AMP-activated protein kinase phosphorylates and
desensitizes smooth muscle myosin light chain kinase. J Biol Chem
283: 18505–18512, 2008.
Huang X, Wullschleger S, Shpiro N, McGuire VA, Sakamoto K,
Woods YL, McBurnie W, Fleming S, Alessi DR. Important role
of the LKB1-AMPK pathway in suppressing tumourigenesis in
PTEN deficient mice. Biochem J 412: 211–221, 2008.
Hurov JB, Huang M, White LS, Lennerz J, Choi CS, Cho YR,
Kim HJ, Prior JL, Piwnica-Worms D, Cantley LC, Kim JK,
Shulman GI, Piwnica-Worms H. Loss of the Par-1b/MARK2 polarity kinase leads to increased metabolic rate, decreased adiposity, and insulin hypersensitivity in vivo. Proc Natl Acad Sci USA
104: 5680 –5685, 2007.
Hurov JB, Watkins JL, Piwnica-Worms H. Atypical PKC phosphorylates PAR-1 kinases to regulate localization and activity. Curr
Biol 14: 736 –741, 2004.
Jang C, Lee G, Chung J. LKB1 induces apical trafficking of
Silnoon, a monocarboxylate transporter, in Drosophila melanogaster. J Cell Biol 183: 11–17, 2008.
Jansen M, de Leng WW, Baas AF, Myoshi H, Mathus-Vliegen
L, Taketo MM, Clevers H, Giardiello FM, Offerhaus GJ.
Mucosal prolapse in the pathogenesis of Peutz-Jeghers polyposis.
Gut 55: 1–5, 2006.
Jenne DE, Reimann H, Nezu J, Friedel W, Loff S, Jeschke R,
Muller O, Back W, Zimmer M. Peutz-Jeghers syndrome is caused
by mutations in a novel serine threonine kinase. Nature Genet 18:
38 – 43, 1998.
Ji H, Ramsey MR, Hayes DN, Fan C, McNamara K, Kozlowski
P, Torrice C, Wu MC, Shimamura T, Perera SA, Liang MC, Cai
D, Naumov GN, Bao L, Contreras CM, Li D, Chen L, Krishnamurthy J, Koivunen J, Chirieac LR, Padera RF, Bronson
RT, Lindeman NI, Christiani DC, Lin X, Shapiro GI, Janne PA,
Johnson BE, Meyerson M, Kwiatkowski DJ, Castrillon DH,
Bardeesy N, Sharpless NE, Wong KK. LKB1 modulates lung
cancer differentiation and metastasis. Nature 448: 807– 810, 2007.
Jishage K, Nezu J, Kawase Y, Iwata T, Watanabe M, Miyoshi
A, Ose A, Habu K, Kake T, Kamada N, Ueda O, Kinoshita M,
Jenne DE, Shimane M, Suzuki H. Role of Lkb1, the causative
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
22. Brajenovic M, Joberty G, Kuster B, Bouwmeester T, Drewes
G. Comprehensive proteomic analysis of human Par protein complexes reveals an interconnected protein network. J Biol Chem
279: 12804 –12811, 2004.
23. Brosens LA, van Hattem WA, Jansen M, de Leng WW, Giardiello FM, Offerhaus GJ. Gastrointestinal polyposis syndromes. Curr Mol Med 7: 29 – 46, 2007.
24. Brugarolas J, Kaelin WG Jr. Dysregulation of HIF and VEGF is
a unifying feature of the familial hamartoma syndromes. Cancer
Cell 6: 7–10, 2004.
25. Chen YM, Wang QJ, Hu HS, Yu PC, Zhu J, Drewes G, PiwnicaWorms H, Luo ZG. Microtubule affinity-regulating kinase 2 functions downstream of the PAR-3/PAR-6/atypical PKC complex in
regulating hippocampal neuronal polarity. Proc Natl Acad Sci USA
103: 8534 – 8539, 2006.
26. Collins SP, Reoma JL, Gamm DM, Uhler MD. LKB1, a novel
serine/threonine protein kinase and potential tumour suppressor, is
phosphorylated by cAMP-dependent protein kinase (PKA) and prenylated in vivo. Biochem J 345: 673– 680, 2000.
27. Contreras CM, Gurumurthy S, Haynie JM, Shirley LJ, Akbay
EA, Wingo SN, Schorge JO, Broaddus RR, Wong KK, Bardeesy
N, Castrillon DH. Loss of Lkb1 provokes highly invasive endometrial adenocarcinomas. Cancer Res 68: 759 –766, 2008.
28. Cowan CR, Hyman AA. Acto-myosin reorganization and PAR
polarity in C. elegans. Development 134: 1035–1043, 2007.
29. De Leng WW, Jansen M, Carvalho R, Polak M, Musler AR,
Milne AN, Keller JJ, Menko FH, de Rooij FW, IacobuzioDonahue CA, Giardiello FM, Weterman MA, Offerhaus GJ.
Genetic defects underlying Peutz-Jeghers syndrome (PJS) and exclusion of the polarity-associated MARK/Par1 gene family as potential PJS candidates. Clin Genet 72: 568 –573, 2007.
30. De Leng WW, Jansen M, Keller JJ, de Gijsel M, Milne AN,
Morsink FH, Weterman MA, Iacobuzio-Donahue CA, Clevers
HC, Giardiello FM, Offerhaus GJ. Peutz-Jeghers syndrome polyps are polyclonal with expanded progenitor cell compartment.
Gut 56: 1475–1476, 2007.
31. De Leng WW, Keller JJ, Luiten S, Musler AR, Jansen M, Baas
AF, de Rooij FW, Gille JJ, Menko FH, Offerhaus GJ, Weterman MA. STRAD in Peutz-Jeghers syndrome and sporadic cancers.
J Clin Pathol 58: 1091–1095, 2005.
32. Denison FC, Hiscock NJ, Carling D, Woods A. Characterization
of an alternative splice variant of LKB1. J Biol Chem 284: 67–76,
2009.
33. Dorfman J, Macara IG. STRAD␣ regulates LKB1 localization by
blocking access to importin-␣, and by association with Crm1 and
Exportin-7. Mol Biol Cell 19: 1614 –1626, 2008.
34. Drewes G, Ebneth A, Mandelkow EM. MAPs, MARKs and microtubule dynamics. Trends Biochem Sci 23: 307–311, 1998.
35. Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris
AD. Metformin and reduced risk of cancer in diabetic patients.
BMJ 330: 1304 –1305, 2005.
36. Giardiello FM, Trimbath JD. Peutz-Jeghers syndrome and management recommendations. Clin Gastroenterol Hepatol 4: 408 –
415, 2006.
37. Goldstein B, Macara IG. The PAR proteins: fundamental players
in animal cell polarization. Dev Cell 13: 609 – 622, 2007.
38. Goransson O, Deak M, Wullschleger S, Morrice NA, Prescott
AR, Alessi DR. Regulation of the polarity kinases PAR-1/MARK by
14-3-3 interaction and phosphorylation. J Cell Sci 119: 4059 – 4070,
2006.
39. Gurumurthy S, Hezel AF, Berger JH, Bosenberg MW, Bardeesy N. LKB1 deficiency sensitizes mice to carcinogen-induced
tumorigenesis. Cancer Res 68: 55– 63, 2008.
40. Gwinn DM, Shackelford DB, Egan DF, Mihaylova MM, Mery
A, Vasquez DS, Turk BE, Shaw RJ. AMPK phosphorylation of
raptor mediates a metabolic checkpoint. Mol Cell 30: 214 –226,
2008.
41. Hardie DG. AMP-activated/SNF1 protein kinases: conserved guardians of cellular energy. Nature Rev 8: 774 –785, 2007.
42. Hawley SA, Boudeau J, Reid JL, Mustard KJ, Udd L, Makela
TP, Alessi DR, Hardie DG. Complexes between the LKB1 tumor
suppressor, STRAD alpha/beta and MO25 alpha/beta are upstream
795
796
58.
59.
60.
61.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
gene of Peutz-Jegher’s syndrome, in embryogenesis and polyposis.
Proc Natl Acad Sci USA 99: 8903– 8908, 2002.
Jorgensen SB, Viollet B, Andreelli F, Frosig C, Birk JB,
Schjerling P, Vaulont S, Richter EA, Wojtaszewski JF. Knockout of the alpha2 but not alpha1 5’-AMP-activated protein kinase
isoform abolishes 5-aminoimidazole-4-carboxamide-1-beta-4-ribofuranoside but not contraction-induced glucose uptake in skeletal
muscle. J Biol Chem 279: 1070 –1079, 2004.
Katajisto P, Vaahtomeri K, Ekman N, Ventela E, Ristimaki A,
Bardeesy N, Feil R, DePinho RA, Makela TP. LKB1 signaling in
mesenchymal cells required for suppression of gastrointestinal
polyposis. Nature Genet 40: 455– 459, 2008.
Katajisto P, Vallenius T, Vaahtomeri K, Ekman N, Udd L,
Tiainen M, Makela TP. The LKB1 tumor suppressor kinase in
human disease. Biochim Biophys Acta 1775: 63–75, 2007.
Keller JJ, Westerman AM, de Rooij FW, Wilson JH, van
Dekken H, Giardiello FM, Weterman MA, Offerhaus GJ. Molecular genetic evidence of an association between nasal polyposis
and the Peutz-Jeghers syndrome. Ann Int Med 136: 855– 856, 2002.
Kemphues KJ, Priess JR, Morton DG, Cheng NS. Identification
of genes required for cytoplasmic localization in early C. elegans
embryos. Cell 52: 311–320, 1988.
Kim KM, Calabrese P, Tavare S, Shibata D. Enhanced stem cell
survival in familial adenomatous polyposis. Am J Pathol 164: 1369 –
1377, 2004.
Kinzler KW, Vogelstein B. Landscaping the cancer terrain. Science 280: 1036 –1037, 1998.
Kishi M, Pan YA, Crump JG, Sanes JR. Mammalian SAD kinases
are required for neuronal polarization. Science 307: 929 –932, 2005.
Klezovitch O, Fernandez TE, Tapscott SJ, Vasioukhin V. Loss
of cell polarity causes severe brain dysplasia in Lgl1 knockout
mice. Genes Dev 18: 559 –571, 2004.
Knoblich JA. Mechanisms of asymmetric stem cell division. Cell
132: 583–597, 2008.
Koh HJ, Arnolds DE, Fujii N, Tran TT, Rogers MJ, Jessen N,
Li Y, Liew CW, Ho RC, Hirshman MF, Kulkarni RN, Kahn CR,
Goodyear LJ. Skeletal muscle-selective knockout of LKB1 increases insulin sensitivity, improves glucose homeostasis, and decreases TRB3. Mol Cell Biol 26: 8217– 8227, 2006.
Kojima Y, Miyoshi H, Clevers HC, Oshima M, Aoki M, Taketo
MM. Suppression of tubulin polymerization by the LKB1-microtubule-associated protein/microtubule affinity-regulating kinase signaling. J Biol Chem 282: 23532–23540, 2007.
Lechler T, Fuchs E. Asymmetric cell divisions promote stratification and differentiation of mammalian skin. Nature 437: 275–280,
2005.
Lee CY, Robinson KJ, Doe CQ. Lgl, Pins and aPKC regulate
neuroblast self-renewal versus differentiation. Nature 439: 594 –
598, 2006.
Lee JH, Koh H, Kim M, Kim Y, Lee SY, Karess RE, Lee SH,
Shong M, Kim JM, Kim J, Chung J. Energy-dependent regulation
of cell structure by AMP-activated protein kinase. Nature 447:
1017–1020, 2007.
Leedham SJ, Wright NA. Expansion of a mutated clone: from
stem cell to tumour. J Clin Pathol 61: 164 –171, 2008.
Lim W, Hearle N, Shah B, Murday V, Hodgson SV, Lucassen A,
Eccles D, Talbot I, Neale K, Lim AG, O’Donohue J, Donaldson
A, Macdonald RC, Young ID, Robinson MH, Lee PW, Stoodley
BJ, Tomlinson I, Alderson D, Holbrook AG, Vyas S, Swarbrick
ET, Lewis AA, Phillips RK, Houlston RS. Further observations
on LKB1/STK11 status and cancer risk in Peutz-Jeghers syndrome.
Br J Cancer 89: 308 –313, 2003.
Linos DA, Dozois RR, Dahlin DC, Bartholomew LG. Does
Peutz-Jeghers syndrome predispose to gastrointestinal malignancy? A later look. Arch Surg 116: 1182–1184, 1981.
Lizcano JM, Goransson O, Toth R, Deak M, Morrice NA,
Boudeau J, Hawley SA, Udd L, Makela TP, Hardie DG, Alessi
DR. LKB1 is a master kinase that activates 13 kinases of the AMPK
subfamily, including MARK/PAR-1. EMBO J 23: 833– 843, 2004.
Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S.
The protein kinase complement of the human genome. Science 298:
1912–1934, 2002.
Physiol Rev • VOL
78. Martin SG, St Johnston D. A role for Drosophila LKB1 in anterior-posterior axis formation and epithelial polarity. Nature 421:
379 –384, 2003.
79. Matsumoto S, Iwakawa R, Takahashi K, Kohno T, Nakanishi
Y, Matsuno Y, Suzuki K, Nakamoto M, Shimizu E, Minna JD,
Yokota J. Prevalence and specificity of LKB1 genetic alterations in
lung cancers. Oncogene 26: 5911–5918, 2007.
80. McGarrity TJ, Amos C. Peutz-Jeghers syndrome: clinicopathology and molecular alterations. Cell Mol Life Sci 63: 2135–2144,
2006.
81. McGee SL, Mustard KJ, Hardie DG, Baar K. Normal hypertrophy accompanied by phosphoryation and activation of AMP-activated protein kinase alpha1 following overload in LKB1 knockout
mice. J Physiol 586: 1731–1741, 2008.
82. Milburn CC, Boudeau J, Deak M, Alessi DR, van Aalten DM.
Crystal structure of MO25 alpha in complex with the C terminus of
the pseudo kinase STE20-related adaptor. Nature Struct Mol Biol
11: 193–200, 2004.
83. Miller LJ, Bartholomew LG, Dozois RR, Dahlin DC. Adenocarcinoma of the rectum arising in a hamartomatous polyp in a patient
with Peutz-Jeghers syndrome. Dig Dis Sci 28: 1047–1051, 1983.
84. Mirouse V, Swick LL, Kazgan N, St Johnston D, Brenman JE.
LKB1 and AMPK maintain epithelial cell polarity under energetic
stress. J Cell Biol 177: 387–392, 2007.
85. Miyoshi H, Nakau M, Ishikawa TO, Seldin MF, Oshima M,
Taketo MM. Gastrointestinal hamartomatous polyposis in Lkb1
heterozygous knockout mice. Cancer Res 62: 2261–2266, 2002.
86. Morton DG, Roos JM, Kemphues KJ. par-4, a gene required for
cytoplasmic localization and determination of specific cell types in
Caenorhabditis elegans embryogenesis. Genetics 130: 771–790,
1992.
87. Nakau M, Miyoshi H, Seldin MF, Imamura M, Oshima M,
Taketo MM. Hepatocellular carcinoma caused by loss of heterozygosity in Lkb1 gene knockout mice. Cancer Res 62: 4549 – 4553,
2002.
88. Narkar VA, Downes M, Yu RT, Embler E, Wang YX, Banayo E,
Mihaylova MM, Nelson MC, Zou Y, Juguilon H, Kang H, Shaw
RJ, Evans RM. AMPK and PPARdelta agonists are exercise mimetics. Cell 134: 405– 415, 2008.
89. Parsons DW, Wang TL, Samuels Y, Bardelli A, Cummins JM,
DeLong L, Silliman N, Ptak J, Szabo S, Willson JK, Markowitz
S, Kinzler KW, Vogelstein B, Lengauer C, Velculescu VE.
Colorectal cancer: mutations in a signalling pathway. Nature 436:
792, 2005.
90. Pearson HB, McCarthy A, Collins CM, Ashworth A, Clarke
AR. Lkb1 deficiency causes prostate neoplasia in the mouse. Cancer Res 68: 2223–2232, 2008.
91. Puffenberger EG, Strauss KA, Ramsey KE, Craig DW,
Stephan DA, Robinson DL, Hendrickson CL, Gottlieb S, Ramsay DA, Siu VM, Heuer GG, Crino PB, Morton DH. Polyhydramnios, megalencephaly and symptomatic epilepsy caused by a homozygous 7-kilobase deletion in LYK5. Brain 130: 1929 –1941, 2007.
92. Rossi DJ, Ylikorkala A, Korsisaari N, Salovaara R, Luukko K,
Launonen V, Henkemeyer M, Ristimaki A, Aaltonen LA,
Makela TP. Induction of cyclooxygenase-2 in a mouse model of
Peutz-Jeghers polyposis. Proc Natl Acad Sci USA 99: 12327–12332,
2002.
93. Rustgi AK. The genetics of hereditary colon cancer. Genes Dev 21:
2525–2538, 2007.
94. Sakamoto K, Goransson O, Hardie DG, Alessi DR. Activity of
LKB1 and AMPK-related kinases in skeletal muscle: effects of
contraction, phenformin, and AICAR. Am J Physiol Endocrinol
Metab 287: E310 –E317, 2004.
95. Sakamoto K, McCarthy A, Smith D, Green KA, Grahame Hardie D, Ashworth A, Alessi DR. Deficiency of LKB1 in skeletal
muscle prevents AMPK activation and glucose uptake during contraction. EMBO J 24: 1810 –1820, 2005.
96. Salmena L, Carracedo A, Pandolfi PP. Tenets of PTEN tumor
suppression. Cell 133: 403– 414, 2008.
97. Sanchez-Cespedes M, Parrella P, Esteller M, Nomoto S, Trink
B, Engles JM, Westra WH, Herman JG, Sidransky D. Inactivation of LKB1/STK11 is a common event in adenocarcinomas of the
lung. Cancer Res 62: 3659 –3662, 2002.
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
62.
JANSEN ET AL.
LKB1-AMPK FAMILY SIGNALING
Physiol Rev • VOL
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
tumor cell anabolism and suppresses the malignant phenotype.
Cancer Res 65: 2441–2448, 2005.
Takeda H, Miyoshi H, Kojima Y, Oshima M, Taketo MM. Accelerated onsets of gastric hamartomas and hepatic adenomas/
carcinomas in Lkb1⫹/⫺p53⫺/⫺ compound mutant mice. Oncogene
25: 1816 –1820, 2006.
Thomson DM, Brown JD, Fillmore N, Condon BM, Kim HJ,
Barrow JR, Winder WW. LKB1 and the regulation of malonyl-CoA
and fatty acid oxidation in muscle. Am J Physiol Endocrinol Metab
293: E1572–E1579, 2007.
Tian AG, Deng WM. Lgl and its phosphorylation by aPKC regulate
oocyte polarity formation in Drosophila. Development 135: 463–
471, 2008.
Timm T, Li XY, Biernat J, Jiao J, Mandelkow E, Vandekerckhove J, Mandelkow EM. MARKK, a Ste20-like kinase, activates
the polarity-inducing kinase MARK/PAR-1. EMBO J 22: 5090 –5101,
2003.
Towler MC, Fogarty S, Hawley SA, Pan DA, Martin DM, Morrice NA, McCarthy A, Galardo MN, Meroni SB, Cigorraga SB,
Ashworth A, Sakamoto K, Hardie DG. A novel short splice
variant of the tumour suppressor LKB1 is required for spermiogenesis. Biochem J 416: 1–14, 2008.
Towler MC, Hardie DG. AMP-activated protein kinase in metabolic control and insulin signaling. Circ Res 100: 328 –341, 2007.
Trojan J, Brieger A, Raedle J, Esteller M, Zeuzem S. 5⬘-CpG
island methylation of the LKB1/STK11 promoter and allelic loss at
chromosome 19p13.3 in sporadic colorectal cancer. Gut 47: 272–
276, 2000.
Udd L, Katajisto P, Rossi DJ, Lepisto A, Lahesmaa AM,
Ylikorkala A, Jarvinen HJ, Ristimaki AP, Makela TP. Suppression of Peutz-Jeghers polyposis by inhibition of cyclooxygenase-2.
Gastroenterology 127: 1030 –1037, 2004.
Vaahtomeri K, Ventela E, Laajanen K, Katajisto P, Wipff PJ,
Hinz B, Vallenius T, Tiainen M, Makela TP. Lkb1 is required for
TGF␤-mediated myofibroblast differentiation. J Cell Sci 121: 3531–
3540, 2008.
Viollet B, Andreelli F, Jorgensen SB, Perrin C, Geloen A,
Flamez D, Mu J, Lenzner C, Baud O, Bennoun M, Gomas E,
Nicolas G, Wojtaszewski JF, Kahn A, Carling D, Schuit FC,
Birnbaum MJ, Richter EA, Burcelin R, Vaulont S. The AMPactivated protein kinase alpha2 catalytic subunit controls wholebody insulin sensitivity. J Clin Invest 111: 91–98, 2003.
Watts JL, Morton DG, Bestman J, Kemphues KJ. The C.
elegans par-4 gene encodes a putative serine-threonine kinase
required for establishing embryonic asymmetry. Development 127:
1467–1475, 2000.
Wei C, Amos CI, Stephens LC, Campos I, Deng JM, Behringer
RR, Rashid A, Frazier ML. Mutation of Lkb1 and p53 genes exert
a cooperative effect on tumorigenesis. Cancer Res 65: 11297–11303,
2005.
Wei C, Amos CI, Zhang N, Wang X, Rashid A, Walker CL,
Behringer RR, Frazier ML. Suppression of Peutz-Jeghers
polyposis by targeting mammalian target of rapamycin signaling.
Clin Cancer Res 14: 1167–1171, 2008.
Westerman AM, Entius MM, de Baar E, Boor PP, Koole R, van
Velthuysen ML, Offerhaus GJ, Lindhout D, de Rooij FW, Wilson JH. Peutz-Jeghers syndrome: 78-year follow-up of the original
family. Lancet 353: 1211–1215, 1999.
Winckler B. BDNF instructs the kinase LKB1 to grow an axon. Cell
129: 459 – 460, 2007.
Woo A, Sadana A, Mauger DT, Baker MJ, Berk T, McGarrity
TJ. Psychosocial impact of Peutz-Jeghers Syndrome. Fam Cancer
8: 59 – 65, 2009.
Woods A, Johnstone SR, Dickerson K, Leiper FC, Fryer LG,
Neumann D, Schlattner U, Wallimann T, Carlson M, Carling
D. LKB1 is the upstream kinase in the AMP-activated protein
kinase cascade. Curr Biol 13: 2004 –2008, 2003.
Yamamoto Y, Izumi Y, Matsuzaki F. The GC kinase Fray and
Mo25 regulate Drosophila asymmetric divisions. Biochem Biophys
Res Commun 366: 212–218, 2008.
Ylikorkala A, Rossi DJ, Korsisaari N, Luukko K, Alitalo K,
Henkemeyer M, Makela TP. Vascular abnormalities and deregu-
89 • JULY 2009 •
www.prv.org
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
98. Sanders MJ, Grondin PO, Hegarty BD, Snowden MA, Carling
D. Investigating the mechanism for AMP activation of the AMPactivated protein kinase cascade. Biochem J 403: 139 –148, 2007.
99. Sapkota GP, Deak M, Kieloch A, Morrice N, Goodarzi AA,
Smythe C, Shiloh Y, Lees-Miller SP, Alessi DR. Ionizing radiation induces ataxia telangiectasia mutated kinase (ATM)-mediated
phosphorylation of LKB1/STK11 at Thr-366. Biochem J 368: 507–
516, 2002.
100. Sapkota GP, Kieloch A, Lizcano JM, Lain S, Arthur JS,
Williams MR, Morrice N, Deak M, Alessi DR. Phosphorylation of the protein kinase mutated in Peutz-Jeghers cancer syndrome, LKB1/STK11, at Ser431 by p90(RSK) and cAMPdependent protein kinase, but not its farnesylation at Cys(433),
is essential for LKB1 to suppress cell growth. J Biol Chem 276:
19469 –19482, 2001.
101. Sato N, Rosty C, Jansen M, Fukushima N, Ueki T, Yeo CJ,
Cameron JL, Iacobuzio-Donahue CA, Hruban RH, Goggins M.
STK11/LKB1 Peutz-Jeghers gene inactivation in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol 159: 2017–
2022, 2001.
102. Shah U, Sharpless NE, Hayes DN. LKB1 and lung cancer: more
than the usual suspects. Cancer Res 68: 3562–3565, 2008.
103. Shaw RJ, Bardeesy N, Manning BD, Lopez L, Kosmatka M,
DePinho RA, Cantley LC. The LKB1 tumor suppressor negatively
regulates mTOR signaling. Cancer Cell 6: 91–99, 2004.
104. Shaw RJ, Kosmatka M, Bardeesy N, Hurley RL, Witters LA,
DePinho RA, Cantley LC. The tumor suppressor LKB1 kinase
directly activates AMP-activated kinase and regulates apoptosis in
response to energy stress. Proc Natl Acad Sci USA 101: 3329 –3335,
2004.
105. Shaw RJ, Lamia KA, Vasquez D, Koo SH, Bardeesy N, Depinho
RA, Montminy M, Cantley LC. The kinase LKB1 mediates glucose homeostasis in liver and therapeutic effects of metformin.
Science 310: 1642–1646, 2005.
106. Shepherd NA, Bussey HJ, Jass JR. Epithelial misplacement in
Peutz-Jeghers polyps. A diagnostic pitfall. Am J Surg Pathol 11:
743–749, 1987.
107. Singh B, Mortensen NJ, Warren BF. Histopathological mimicry
in mucosal prolapse. Histopathology 50: 97–102, 2007.
108. Smith DP, Spicer J, Smith A, Swift S, Ashworth A. The mouse
Peutz-Jeghers syndrome gene Lkb1 encodes a nuclear protein kinase. Hum Mol Genet 8: 1479 –1485, 1999.
109. Song P, Xie Z, Wu Y, Xu J, Dong Y, Zou MH. Protein kinase C
zeta-dependent LKB1 serine 428 phosphorylation increases LKB1
nucleus export and apoptosis in endothelial cells. J Biol Chem 283:
12446 –12455, 2008.
110. Spicer J, Rayter S, Young N, Elliott R, Ashworth A, Smith
D. Regulation of the Wnt signalling component PAR1A by the
Peutz-Jeghers syndrome kinase LKB1. Oncogene 22: 4752– 4756,
2003.
111. Stapleton D, Mitchelhill KI, Gao G, Widmer J, Michell BJ, Teh
T, House CM, Fernandez CS, Cox T, Witters LA, Kemp BE.
Mammalian AMP-activated protein kinase subfamily. J Biol Chem
271: 611– 614, 1996.
112. Suter M, Riek U, Tuerk R, Schlattner U, Wallimann T, Neumann D. Dissecting the role of 5⬘-AMP for allosteric stimulation,
activation, and deactivation of AMP-activated protein kinase. J Biol
Chem 281: 32207–32216, 2006.
113. Suzuki A, Hirata M, Kamimura K, Maniwa R, Yamanaka T,
Mizuno K, Kishikawa M, Hirose H, Amano Y, Izumi N, Miwa Y,
Ohno S. aPKC acts upstream of PAR-1b in both the establishment
and maintenance of mammalian epithelial polarity. Curr Biol 14:
1425–1435, 2004.
114. Suzuki A, Okamoto S, Lee S, Saito K, Shiuchi T, Minokoshi Y.
Leptin stimulates fatty acid oxidation and peroxisome proliferatoractivated receptor alpha gene expression in mouse C2C12 myoblasts by changing the subcellular localization of the alpha2 form of
AMP-activated protein kinase. Mol Cell Biol 27: 4317– 4327, 2007.
115. Swinnen JV, Beckers A, Brusselmans K, Organe S, Segers J,
Timmermans L, Vanderhoydonc F, Deboel L, Derua R, Waelkens E, De Schrijver E, Van de Sande T, Noel A, Foufelle F,
Verhoeven G. Mimicry of a cellular low energy status blocks
797
798
JANSEN ET AL.
lation of VEGF in Lkb1-deficient mice. Science 293: 1323–1326,
2001.
135. Young RH. Sex cord-stromal tumors of the ovary and testis: their
similarities and differences with consideration of selected problems. Mod Pathol 18 Suppl 2: S81–98, 2005.
136. Zhang L, Li J, Young LH, Caplan MJ. AMP-activated protein
kinase regulates the assembly of epithelial tight junctions. Proc
Natl Acad Sci USA 103: 17272–17277, 2006.
137. Zhang S, Schafer-Hales K, Khuri FR, Zhou W, Vertino PM, Marcus AI. The tumor suppressor LKB1 regulates lung cancer cell polar-
ity by mediating cdc42 recruitment and activity. Cancer Res 68: 740 –
748, 2008.
138. Zheng B, Cantley LC. Regulation of epithelial tight junction assembly and disassembly by AMP-activated protein kinase. Proc
Natl Acad Sci USA 104: 819 – 822, 2007.
139. Zhou G, Myers R, Li Y, Chen Y, Shen X, Fenyk-Melody J, Wu
M, Ventre J, Doebber T, Fujii N, Musi N, Hirshman MF,
Goodyear LJ, Moller DE. Role of AMP-activated protein kinase in
mechanism of metformin action. J Clin Invest 108: 1167–1174,
2001.
Downloaded from http://physrev.physiology.org/ by 10.220.33.6 on July 4, 2017
Physiol Rev • VOL
89 • JULY 2009 •
www.prv.org